timeline.json

{"users": {"kulcheema": {"id": "100357360", "screen_name": "kulcheema", "name": "Kulpreet Cheema \ud83e\udde0\u00a0\ud83d\udcd6 \ud83d\udcbb", "avatar": "https://pbs.twimg.com/profile_images/1146880679663173632/vOSJWmKY_normal.png", "description": "@NSERC_CRSNG Neuroscience PhD student at @UAlberta_FoMD, Interested in functional brain connectivity, #scicomm #neuroimaging"}, "IdanAsherBlank": {"id": "4764909501", "screen_name": "IdanAsherBlank", "name": "Idan Blank", "avatar": "https://pbs.twimg.com/profile_images/914958386541862913/F9C1XVf2_normal.jpg", "description": "Cognitive (neuro)scientist studying language at UCLA Psych (BlankLangLab). Views are my own. He/Him. RT \u2260 endorsement. \ud83c\udff3\ufe0f\u200d\ud83c\udf08"}, "MaggieMaeWokeUp": {"id": "101262586", "screen_name": "MaggieMaeWokeUp", "name": "Maggie Mae Mell", "avatar": "https://pbs.twimg.com/profile_images/1085185186952814593/kwCCoCZs_normal.jpg", "description": "PhD Candidate in visual computational neuroscience. She/Her \ud83d\udc41\ud83e\udde0\ud83d\udc69\ud83c\udffb\u200d\ud83d\udd2c\ud83d\udc69\ud83c\udffc\u200d\ud83c\udfa4  Opinions my own and do not represent any org I am affiliated with."}, "crntozlu": {"id": "2428607799", "screen_name": "crntozlu", "name": "Ceren Tozlu", "avatar": "https://pbs.twimg.com/profile_images/494136064811237376/BZ9nAFBe_normal.jpeg", "description": "Postdoc in the Department of Radiology at Weill Cornell Medicine -  CoCo Lab"}, "MarcCoutanche": {"id": "3611410942", "screen_name": "MarcCoutanche", "name": "Marc Coutanche", "avatar": "https://pbs.twimg.com/profile_images/1158194251295076353/-XGP_We4_normal.jpg", "description": "Assistant Prof of Psychology @ U. of Pittsburgh. Cognitive neuroscience, memory, learning, perception, fMRI pattern analysis. Born in \ud83c\uddef\ud83c\uddea, Live in \ud83c\uddfa\ud83c\uddf8. He/him"}, "jpeelle": {"id": "37563", "screen_name": "jpeelle", "name": "Jonathan Peelle", "avatar": "https://pbs.twimg.com/profile_images/1649546185/photo_small_normal.jpg", "description": "Looking for meaningful (mostly auditory) things in the human brain \u2022 science, inclusiveness. #academicparent \u2022 he/him\n\nCo-host @juicesqueezepod\n\nLab: @peellelab"}, "DavideValeriani": {"id": "54594638", "screen_name": "DavideValeriani", "name": "Davide Valeriani, PhD", "avatar": "https://pbs.twimg.com/profile_images/1240660514461622276/Gss36Yx1_normal.jpg", "description": "Decision making + neuroimaging + machine learning + BCI. Research fellow at Harvard @SimonyanLab. Maintainer of @matplotlib's blog, @harvardmpa board."}, "elvisha9": {"id": "1233507200183787525", "screen_name": "elvisha9", "name": "Elvisha Dhamala", "avatar": "https://pbs.twimg.com/profile_images/1233507330307903496/Vm993mFW_normal.jpg", "description": "Neuroscience Ph.D. Candidate @WeillCornellRad | Social Coordinator Elect @OHBM_Trainees | Instructor @Cornell Prison Education Program"}, "OHBM_Trainees": {"id": "3236734336", "screen_name": "OHBM_Trainees", "name": "OHBM Trainees", "avatar": "https://pbs.twimg.com/profile_images/996156451897987072/nQd4KoaJ_normal.jpg", "description": "Organization for Human Brain Mapping - Student and Postdoc Special Interest Group. Tweets by @amhaugg @krsitek"}, "BIDSstandard": {"id": "1133451131886030848", "screen_name": "BIDSstandard", "name": "BIDS-standard", "avatar": "https://pbs.twimg.com/profile_images/1136717546881552384/2ZZcKmyl_normal.png", "description": "Brain Imaging Data Structure. Simple and intuitive way to organize and describe your neuroimaging and behavioral data. https://t.co/u2BDPDmmxa"}, "JohnSamuelsson1": {"id": "1159676473273454592", "screen_name": "JohnSamuelsson1", "name": "John Samuelsson", "avatar": "https://pbs.twimg.com/profile_images/1159680497364164609/1aL_9ISV_normal.jpg", "description": ""}, "Aman_Badhwar": {"id": "3191006281", "screen_name": "Aman_Badhwar", "name": "AmanPreet Badhwar", "avatar": "https://pbs.twimg.com/profile_images/597440133676183552/MMyDAXey_normal.jpg", "description": "@CIHR & ASRP Postdoc (Alzheimer, Multiomics); Artist; President@CCNA_Trainees; Chair@OHBM Brain-Art; Chair-elect@OHBM Diversity; an editor@OHBM blog; She/Her/Dr"}, "matteo_brainnet": {"id": "263807286", "screen_name": "matteo_brainnet", "name": "Matteo Fraschini", "avatar": "https://pbs.twimg.com/profile_images/1225004523858595840/tiVQ1_fM_normal.jpg", "description": "Assistant Professor of Computer Engineering @univca interested in M/EEG Brain Networks. Dad. RTs are (generally) bookmarks."}, "UTooley": {"id": "63304474", "screen_name": "UTooley", "name": "Ursula Tooley", "avatar": "https://pbs.twimg.com/profile_images/1063913623922401280/6zhFxXHw_normal.jpg", "description": "Neuroscience Ph.D. candidate at UPenn | dev cog neuro | networks | environment | she/her"}, "LucinaUddin": {"id": "2886972634", "screen_name": "LucinaUddin", "name": "Lucina Uddin", "avatar": "https://pbs.twimg.com/profile_images/651564905720492032/E4p2yQBX_normal.jpg", "description": "Edgy cognitive neuroscientist"}, "bjbalas": {"id": "193510146", "screen_name": "bjbalas", "name": "Ben Balas", "avatar": "https://pbs.twimg.com/profile_images/1181299273473777664/APdUd4o8_normal.jpg", "description": "Vision Scientist, comics fan, retro-gamer, and dad."}, "ChadsRogers": {"id": "1967820541", "screen_name": "ChadsRogers", "name": "Prof Chad Rogers", "avatar": "https://pbs.twimg.com/profile_images/717045087688065024/2TvRAsaS_normal.jpg", "description": "Cog Neuro & Speech Science // Assistant Professor at Union College in NY"}, "CMagnain": {"id": "1111396123971710976", "screen_name": "CMagnain", "name": "Caroline Magnain", "avatar": "https://pbs.twimg.com/profile_images/1111396479724195840/oRqRSU88_normal.jpg", "description": "Physicist looking at the human brain under my microscope"}, "xx_mengjia": {"id": "4231174733", "screen_name": "xx_mengjia", "name": "Mengjia XU", "avatar": "https://pbs.twimg.com/profile_images/1241226078254964737/5GJbrRyb_normal.jpg", "description": ""}, "russpoldrack": {"id": "85605334", "screen_name": "russpoldrack", "name": "Russ Poldrack", "avatar": "https://pbs.twimg.com/profile_images/1597778808/russ1_jpg_normal", "description": "Professor of Psychology at Stanford. Director of the Stanford Center for Reproducible Neuroscience #FlyingLess"}, "vaibhav_3pathi": {"id": "1190775801307779072", "screen_name": "vaibhav_3pathi", "name": "Vaibhav Tripathi", "avatar": "https://pbs.twimg.com/profile_images/1190775929100021760/MG7XQNci_normal.jpg", "description": "Neuroscientist in training, Engineer\nMeditator, Life long student"}, "AkiNikolaidis": {"id": "308272243", "screen_name": "AkiNikolaidis", "name": "Aki Nikolaidis", "avatar": "https://pbs.twimg.com/profile_images/1009985934027194374/IugtrYlO_normal.jpg", "description": "Neuroscientist interested in MRI, skill learning, AI, brain development, and #OpenScience. He/Him. At: @ChildMindInst @ChildMindRnD @OHBMOpen @OHBM_Trainees"}, "CaseyPaquola": {"id": "978755826830774272", "screen_name": "CaseyPaquola", "name": "Casey Paquola", "avatar": "https://pbs.twimg.com/profile_images/1042722663238651906/6O-3uI8H_normal.jpg", "description": "Postdoctoral fellow in Montreal. Neurocentric"}, "reindervosdwael": {"id": "846902354687418368", "screen_name": "reindervosdwael", "name": "Reinder Vos de Wael", "avatar": "https://pbs.twimg.com/profile_images/924068619742535680/u1HhbYi1_normal.jpg", "description": "Science geek by day, gaming geek by night."}, "GaelVaroquaux": {"id": "370409954", "screen_name": "GaelVaroquaux", "name": "Gael Varoquaux", "avatar": "https://pbs.twimg.com/profile_images/1535103905/gael_normal.png", "description": "Researcher and coder: \u25baData, Brain, & Computational science \u25ba#python / #pydata contributor, (co)founder of scikit-learn & joblib \u25baPhotography on @artgael"}, "vdcalhoun": {"id": "22107961", "screen_name": "vdcalhoun", "name": "Vince Calhoun", "avatar": "https://pbs.twimg.com/profile_images/811983132530348032/azR8VtD-_normal.jpg", "description": "Multimodal brain imaging & genetics, signal & image processing, general all around geek, husband, father of three, lover of purple."}, "AIraji": {"id": "437156990", "screen_name": "AIraji", "name": "Armin Iraji", "avatar": "https://pbs.twimg.com/profile_images/1240014164870844416/-AHqbjN8_normal.jpg", "description": "multimodal brain imaging, machine learning, signal & image processing, neuroscience"}, "pelednoam": {"id": "172762694", "screen_name": "pelednoam", "name": "Noam Peled", "avatar": "https://pbs.twimg.com/profile_images/1161316768297570305/CNTgz1Oh_normal.jpg", "description": ""}, "lauradata": {"id": "2747474142", "screen_name": "lauradata", "name": "Laura Lewis", "avatar": "https://pbs.twimg.com/profile_images/976261147547463680/DVob5dy2_normal.jpg", "description": "neuroscience & imaging, at Boston University"}, "GiuliaBert1": {"id": "1228013784318586883", "screen_name": "GiuliaBert1", "name": "Giulia Bert\u00f2", "avatar": "https://pbs.twimg.com/profile_images/1228013962761003008/jsHUN4f5_normal.jpg", "description": "PhD Candidate @NILabFBK - NeuroInformatics Laboratory - Bruno Kessler Foundation (FBK) - Italy"}, "furranko": {"id": "16105655", "screen_name": "furranko", "name": "Franco Pestilli", "avatar": "https://pbs.twimg.com/profile_images/509117000447049728/5dzdf19s_normal.png", "description": "exploring the scientific #egosystem"}, "T_Hinault": {"id": "791038429714849792", "screen_name": "T_Hinault", "name": "Thomas Hinault", "avatar": "https://pbs.twimg.com/profile_images/1191988310333366272/Jo8LTY6T_normal.png", "description": "Inserm Researcher (U1077).  Neuroscientist and Neuropsychologist"}, "dokatox": {"id": "2602158174", "screen_name": "dokatox", "name": "Dominik Krzemi\u0144ski", "avatar": "https://pbs.twimg.com/profile_images/1088415588903501825/HcVy6o_t_normal.jpg", "description": "PhD student @CUBRICcardiff // Open Source Tech Lead @Appsilon"}, "ettah": {"id": "19587593", "screen_name": "ettah", "name": "Etta Howells", "avatar": "https://pbs.twimg.com/profile_images/1137752259121635329/n-oAFm1C_normal.jpg", "description": "University of Milan. Looking at brain networks using neuroimaging in the neurosurgical setting \ud83c\uddec\ud83c\udde7\ud83c\uddee\ud83c\uddf9"}, "yarenyilmazz": {"id": "772536368266407936", "screen_name": "yarenyilmazz", "name": "Yaren Yilmaz", "avatar": "https://pbs.twimg.com/profile_images/772542850219249665/4OnKMGaQ_normal.jpg", "description": "Medical Student"}, "Sarah_Morgan_UK": {"id": "2538040298", "screen_name": "Sarah_Morgan_UK", "name": "Sarah Morgan", "avatar": "https://pbs.twimg.com/profile_images/771360713600294912/68i-v9nu_normal.jpg", "description": "Theoretical physicist working on brain networks for schizophrenia @ Cambridge Brain Mapping Unit (postdoc). Also excited about diversity in STEM. She/her"}, "saratheriver": {"id": "392322375", "screen_name": "saratheriver", "name": "Sara Lariviere", "avatar": "https://pbs.twimg.com/profile_images/887495290403639297/HYwyfpp2_normal.jpg", "description": "Alternating between being a neuroscience PhD candidate at McGill and a softball beer leaguer tryna hit a home run \ud83e\udd18\ud83c\udffc"}, "BrainImagingUoL": {"id": "1176974867423465472", "screen_name": "BrainImagingUoL", "name": "Keller Neuroimaging Lab", "avatar": "https://pbs.twimg.com/profile_images/1200558839264886786/ZYnmybOt_normal.jpg", "description": "University of Liverpool. Working with @waltoncentre. Structural, functional and metabolic brain imaging biomarkers in health and disease. See also @epilepsy_liv"}, "glassybrain": {"id": "890928444493746176", "screen_name": "glassybrain", "name": "Vanessa Teckentrup", "avatar": "https://pbs.twimg.com/profile_images/941789103976312837/m-oUOGNQ_normal.jpg", "description": "Neuroscience PhD student @neuroMADLAB Brains \u2192 FC \u2192 Stats #OpenScience enthusiast"}, "rubykong92": {"id": "2173729202", "screen_name": "rubykong92", "name": "Ruby Kong", "avatar": "https://pbs.twimg.com/profile_images/1166758098020601856/35W0Gg96_normal.jpg", "description": ""}, "balicea1": {"id": "47962707", "screen_name": "balicea1", "name": "Virtually meeting since 2004", "avatar": "https://pbs.twimg.com/profile_images/1232406809211756545/JkuugICW_normal.png", "description": "Applied deep thinker. Avid empiricist, avocationally representationalist. Director of science and content at @Orthogonal_Lab #WarrenDemocrat"}, "angelatamtweets": {"id": "253297909", "screen_name": "angelatamtweets", "name": "Angela Tam", "avatar": "https://pbs.twimg.com/profile_images/1578669941/DSC00003__2__normal.jpg", "description": "I want to make robot physicians \ud83e\udd16\ud83d\udc89\ud83d\udc8a"}, "fmrib_karla": {"id": "33060093", "screen_name": "fmrib_karla", "name": "Karla Loreen Miller", "avatar": "https://pbs.twimg.com/profile_images/753482077522255872/IYU3OfQi_normal.jpg", "description": "Trained engineer, pretend physicist, neuroscience afficionado. Mostly human. she/her"}, "aneesabrol": {"id": "879852182379544576", "screen_name": "aneesabrol", "name": "Anees Abrol", "avatar": "https://abs.twimg.com/sticky/default_profile_images/default_profile_normal.png", "description": ""}, "deepmilab": {"id": "1233495772903198720", "screen_name": "deepmilab", "name": "DeepMI Lab", "avatar": "https://pbs.twimg.com/profile_images/1236379334325293057/tQJGc0MS_normal.jpg", "description": "AI Methods for Medical Image Analysis - PI Martin Reuter"}, "Fabian_M_Renz": {"id": "1233370060787458054", "screen_name": "Fabian_M_Renz", "name": "Fabian Renz", "avatar": "https://abs.twimg.com/sticky/default_profile_images/default_profile_normal.png", "description": "MSc student - Cognitive Science & Psychology"}, "amhaugg": {"id": "964607535008251904", "screen_name": "amhaugg", "name": "Amelie Haugg", "avatar": "https://pbs.twimg.com/profile_images/1133451170830323714/8qAUIW2X_normal.png", "description": "PhD Student in Neuroscience @ZNZNeuroscience @UZH_science | real-time fMRI #neurofeedback | psychiatric disorders | tweets for @OHBM_Trainees"}, "dr_xenia": {"id": "124834891", "screen_name": "dr_xenia", "name": "Xenia Kobeleva", "avatar": "https://pbs.twimg.com/profile_images/1047498326562787328/YYCebt07_normal.jpg", "description": "Neurologist, researcher, computational neuroscience, medical informatics, brainhacks, digital art, digital medicine, open science, @ University Bonn & DZNE Bonn"}, "sciencebanshee": {"id": "767821356830494720", "screen_name": "sciencebanshee", "name": "Sarah Faber", "avatar": "https://pbs.twimg.com/profile_images/1212722847321395201/2n7l7E3q_normal.jpg", "description": "Screeching about neuroscience, music, G\u00e0idhlig, and the Maritimes. The friendliest mad scientist you'll ever meet. She/her"}, "PVrticka": {"id": "21862422", "screen_name": "PVrticka", "name": "Pascal Vrti\u010dka \ud83e\udde0\ud83d\udc68\u200d\ud83d\udc69\u200d\ud83d\udc66\u200d\ud83d\udc66\ud83e\udd70", "avatar": "https://pbs.twimg.com/profile_images/977493088229822464/QLaO4d4H_normal.jpg", "description": "Lecturer/Assistant Professor @EssexPsychology @Uni_of_Essex. Main focus on the #Social #Neuroscience of #Human #Attachment using #fNIRS, #fMRI and #EEG."}, "R3RT0": {"id": "2231179117", "screen_name": "R3RT0", "name": "Roberto Toro", "avatar": "https://pbs.twimg.com/profile_images/378800000831804003/3bf62c8a34dc87c5b4b7d0e4bda6304a_normal.jpeg", "description": "Into brains (+open science, procedural art, genomics, music...)"}, "katjaQheuer": {"id": "2981013099", "screen_name": "katjaQheuer", "name": "katja heuer", "avatar": "https://pbs.twimg.com/profile_images/556068434593587200/mSg0cjoY_normal.jpeg", "description": "into making, brains, open science, art&science, interactive art, music, javascript, contemporary dance,\nArt of MR https://t.co/HlSSr2gfMQ"}, "ecologician": {"id": "61508887", "screen_name": "ecologician", "name": "Russell Dinnage", "avatar": "https://pbs.twimg.com/profile_images/832377426176913408/E92wu_e7_normal.jpg", "description": "An evolutionary ecologist at the Institute for Applied Ecology, University of Canberra. Biological data science, deep learning, science-inspired generative art"}, "nicolas_traut": {"id": "995938262060593152", "screen_name": "nicolas_traut", "name": "Nicolas Traut", "avatar": "https://pbs.twimg.com/profile_images/1104318795202658306/f29fM15G_normal.png", "description": ""}, "sofievalk": {"id": "112440496", "screen_name": "sofievalk", "name": "Sofie Valk", "avatar": "https://pbs.twimg.com/profile_images/1137077280168169476/4UKdV1bW_normal.jpg", "description": "a very curious cognitive genetic neuroimager | | | | || https://t.co/ZrwWVLtSYB"}, "Lauri_Parkkonen": {"id": "2534841037", "screen_name": "Lauri_Parkkonen", "name": "Lauri Parkkonen", "avatar": "https://pbs.twimg.com/profile_images/855141639274332161/CKeDhQaI_normal.jpg", "description": ""}, "SegoleneGuerin": {"id": "960134309737848833", "screen_name": "SegoleneGuerin", "name": "S\u00e9gol\u00e8ne M. R. Gu\u00e9rin", "avatar": "https://pbs.twimg.com/profile_images/1105036194189053952/M97QwkAR_normal.png", "description": "PhD Student in neurocognitive psychology at SCALab, University of Lille. Interested in motor timing, neurocognitive mecanisms, affective behaviors, fNIRS \ud83e\udde0"}, "BrieucLehmann": {"id": "844485024640589824", "screen_name": "BrieucLehmann", "name": "Brieuc Lehmann", "avatar": "https://pbs.twimg.com/profile_images/976877456660582400/qQuFctpX_normal.jpg", "description": "Postdoc at Big Data Institute, University of Oxford"}, "ABNeuroscience": {"id": "805132486380584961", "screen_name": "ABNeuroscience", "name": "Alex Bowring", "avatar": "https://pbs.twimg.com/profile_images/976413784184184833/AGQ7pcrB_normal.jpg", "description": "Population Health DPhil. at Oxford University. Neuroimaging, functional MRI methods, Reproducibility and Open Science."}, "FPFischmeister": {"id": "1497673452", "screen_name": "FPFischmeister", "name": "Florian Ph.S Fischmeister", "avatar": "https://pbs.twimg.com/profile_images/799616220337147904/v3G4Sxpl_normal.jpg", "description": "Group Leader at @UniGraz, Guest Professor with @AlpenAdriaUni. \nInterested in intervention based plasticity of brain/microbiome, advocating RRI & FAIR principle"}, "f_beissner": {"id": "1239871385822023680", "screen_name": "f_beissner", "name": "Florian Beissner", "avatar": "https://pbs.twimg.com/profile_images/1239871867781107712/jljp3NcO_normal.jpg", "description": "Neuroscientist, Physicist, Endowed Professor of Somatosensory and Autonomic Therapy Research at Hannover Medical School"}, "gerardosalvato": {"id": "169680974", "screen_name": "gerardosalvato", "name": "Gerardo Salvato", "avatar": "https://pbs.twimg.com/profile_images/1116189043254005760/nRpvz3Ox_normal.jpg", "description": "Cognitive Neuroscientist. Postdoc at the University of Pavia and Scientific Consultant at the Niguarda Hospital of Milano. Psychotherapist. Chocolate addicted."}, "wmvanvliet": {"id": "34220287", "screen_name": "wmvanvliet", "name": "Marijn van Vliet", "avatar": "https://pbs.twimg.com/profile_images/1084166159002488835/8m2lnF8J_normal.jpg", "description": "Posthoc postdoc. Neuroscientist specializing in EEG/MEG data analysis and machine learning."}, "robbisg": {"id": "55550359", "screen_name": "robbisg", "name": "Robbis Guidotti", "avatar": "https://pbs.twimg.com/profile_images/2699865531/5923e4fd2a2f88d9335813bc315ce7c7_normal.png", "description": "PostDoc in #Neuroimaging at @univUda, @MamboUnich, now in @AaltoUniversity. I do #BrainDecoding, #ImageProcessing and #MachineLearning with #Python!"}, "Connect2Brain": {"id": "1095980102939222016", "screen_name": "Connect2Brain", "name": "ConnectToBrain", "avatar": "https://pbs.twimg.com/profile_images/1184495665730207744/QNdGmGi1_normal.jpg", "description": "ERC Synergy project (2019-2025) of @AaltoUniversity @uni_tue @univUda developing closed-loop multi-locus transcranial magnetic stimulation. Tweets by @TimoRoine"}, "KhansaRasheed": {"id": "1227459416473178112", "screen_name": "KhansaRasheed", "name": "KHANSA RASHEED", "avatar": "https://pbs.twimg.com/profile_images/1240656967602970630/wtFnQSSP_normal.jpg", "description": "Electrical Engineer\nGraduate Student"}, "FahimiMansoureh": {"id": "952906286064111617", "screen_name": "FahimiMansoureh", "name": "Mansoureh Fahimi", "avatar": "https://pbs.twimg.com/profile_images/1121371864658522112/b23OhMKU_normal.png", "description": ""}, "tschonberg": {"id": "71173089", "screen_name": "tschonberg", "name": "Tom Schonberg", "avatar": "https://pbs.twimg.com/profile_images/1014589232059502593/cYhHYAIG_normal.jpg", "description": "Assistant Prof. in the Department of Neurobiology and Sagol School of Neuroscience @TelAvivUni views are my own #behaviorchange #openscience #brain"}, "LBIAN5": {"id": "1232555368502317056", "screen_name": "LBIAN5", "name": "Lingbin Bian", "avatar": "https://pbs.twimg.com/profile_images/1239873235899961346/e_XqeU3l_normal.jpg", "description": "statistics, computational neuroscience."}, "KGurunandan": {"id": "1233308864835026944", "screen_name": "KGurunandan", "name": "Kshipra Gurunandan", "avatar": "https://pbs.twimg.com/profile_images/1233336748836392960/1WBNQXlr_normal.jpg", "description": "Cognitive neuroscience PhD at @bcbl_ & visiting @BMH_Hub. Studying experience-dependent plasticity of the human brain: structure, function and connectivity."}, "itjohnstone": {"id": "826111302", "screen_name": "itjohnstone", "name": "Tom Johnstone \ud83c\udde6\ud83c\uddfa\ud83c\uddea\ud83c\uddfa", "avatar": "https://pbs.twimg.com/profile_images/1231719273300541441/740c-nXA_normal.jpg", "description": "Director of Neuroimaging, Swinburne University of Technology. Multimodal Imaging, Open Scientific Practice, Emotion, Cognition"}, "pucketta_m": {"id": "753834751056711680", "screen_name": "pucketta_m", "name": "Alexander M. Puckett", "avatar": "https://pbs.twimg.com/profile_images/895911551202283520/oaIu73M5_normal.jpg", "description": ""}, "bttyeo": {"id": "752487999800094720", "screen_name": "bttyeo", "name": "Thomas Yeo", "avatar": "https://pbs.twimg.com/profile_images/864131060237389824/86lpquFH_normal.jpg", "description": "brain imaging, machine learning, mental disorders, big data"}, "kristinasabr": {"id": "715452677434413058", "screen_name": "kristinasabr", "name": "Kristina Sabaroedin", "avatar": "https://pbs.twimg.com/profile_images/1194772949800083456/YN2d8Xcd_normal.jpg", "description": "PhD student at Monash University. Resting-state functional and effective connectivity, psychosis, and dopamine. Also loves yoga, dancing, and art."}, "SidChop": {"id": "777107666585870337", "screen_name": "SidChop", "name": "Sidhant Chopra", "avatar": "https://pbs.twimg.com/profile_images/835014701167718400/-sJQBT3y_normal.jpg", "description": "PhD (ClinPsych) candidate interested in novel ways to improve mental health. Especially interested in clinical neuroscience, psychotherapy and R @BMH_Hub"}, "jmacshine": {"id": "4856065900", "screen_name": "jmacshine", "name": "Mac Shine", "avatar": "https://pbs.twimg.com/profile_images/1017209197732270081/YZBYILGK_normal.jpg", "description": ""}, "AkiZamani": {"id": "2978033358", "screen_name": "AkiZamani", "name": "Akram Zamani", "avatar": "https://pbs.twimg.com/profile_images/1074835725814685696/aj8WAuaY_normal.jpg", "description": "PhD, \ud83e\udde0 postdoctoral Research Fellow @CCSMonash"}, "D_K_Wright": {"id": "1072369798082744320", "screen_name": "D_K_Wright", "name": "David Wright", "avatar": "https://pbs.twimg.com/profile_images/1238680763005976576/PONBNxQ6_normal.jpg", "description": ""}, "cyctbdbw": {"id": "815742722996715520", "screen_name": "cyctbdbw", "name": "Yu-Chi Chen", "avatar": "https://pbs.twimg.com/profile_images/1240847366095114240/let7EBQ2_normal.jpg", "description": "PhD Candidate at @BMH_Hub"}, "LeonieBorne": {"id": "1228144941315936257", "screen_name": "LeonieBorne", "name": "L\u00e9onie Borne", "avatar": "https://pbs.twimg.com/profile_images/1228148078118461440/CHnNjNa0_normal.jpg", "description": "Post-doctoral Research Fellow in Machine Learning and Neuroimaging"}, "NandaRibeiro93": {"id": "920625918804152320", "screen_name": "NandaRibeiro93", "name": "Fernanda Ribeiro", "avatar": "https://pbs.twimg.com/profile_images/978056173487345666/20BtJKlw_normal.jpg", "description": "PhD candidate at Uni of Queensland. In the free time, a yogi wannabe."}, "AshYork1": {"id": "1189395284691836928", "screen_name": "AshYork1", "name": "Ash York", "avatar": "https://pbs.twimg.com/profile_images/1240807968225935360/3JfY8R2x_normal.jpg", "description": "PhD student at UQ, laminar fMRI and cognition. Other loves - 2 dogs and 1 trapeze."}, "csabaorban": {"id": "189611911", "screen_name": "csabaorban", "name": "Csaba Orban", "avatar": "https://pbs.twimg.com/profile_images/1101766420327628801/ftkL8_V__normal.jpg", "description": "neuroscience | physiology | resting state fMRI"}, "DrBreaky": {"id": "3305807792", "screen_name": "DrBreaky", "name": "Michael Breakspear", "avatar": "https://pbs.twimg.com/profile_images/628468793354792960/_3Rx-xjR_normal.jpg", "description": "Neuroscientist, Psychiatrist. Systems neuroscience, computational psychiatry and clinical neuroimaging research."}}, "timeline": [{"date": "2020-03-20T01:51:00+00:00", "text": "TWEET 1/3  The 1st OHBM Equinox Twitter Conference (#OHBMx) starts in a few minutes. Tune in for 24 hours of neuroscience research from across the world! Use the hashtag #OHBMx for posting all the discussions and questions related to the conference.", "media": [], "ids": ["1240818088376823809"], "thread": []}, {"date": "2020-03-20T01:52:00+00:00", "text": "TWEET 2/3 Before we jump in, we would like to take a moment to address the extraordinary situation we all are in due to the coronavirus pandemic. It is important to remember that we all are in this together and for each other as a community. Together we can #flattenthecurve.", "media": [], "ids": ["1240818340379209728"], "thread": []}, {"date": "2020-03-20T01:53:00+00:00", "text": "TWEET 3/3 And finally, a big THANK YOU to all the healthcare professionals who are working incredibly hard in these extremely difficult circumstances. #COVID2019", "media": [], "ids": ["1240818591957729288"], "thread": []}, {"date": "2020-03-20T01:59:00+00:00", "text": "\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\n\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\n\nWelcome to OHBM Equinox 2020 \nSee the program here: https://t.co/BA0kUWU76y\n\n\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013\u2013", "media": [], "ids": ["1240820101647142920"], "thread": []}, {"date": "2020-03-20T02:02:01+00:00", "text": "#OHBMx-1 \u2733 #keynote\n\nMichael Breakspear @drBreaky*\n\n*University of Newcastle\n\n\u25b6 KEYNOTE: Baby brain modes: Towards a \"wave(mode)-particle(network) theory\" of the brain https://t.co/rH0y6RkunV", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThHvVuU8AEvTdb.jpg"}], "ids": ["1240820864083349505"], "thread": [{"id": "1240821100788899840", "user": "DrBreaky", "date": "2020-03-20T02:02:58+00:00", "text": "I\u2019m thrilled to kick-off OHBMx and thank the organizers (@OHBMequinoX  )\n\nThis is a study of neonatal brain health following preterm birth using network theory and modelling with brain modes\n\n  \n\nhttps://t.co/uCApP54yH1", "media": []}, {"id": "1240822597027168256", "user": "DrBreaky", "date": "2020-03-20T02:08:55+00:00", "text": "1. We studied sleep transitions in baby scalp EEG in 2 groups: Preterm (n=42) & full-term (n=52)\n\nAll EEG were acquired at term-equivalent age\n\nQuiet and active sleep in newborn babies are analogues of deep and REM sleep in later life. Their transitions reflect  brain health", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThJevgU4AARcmA.jpg"}]}, {"id": "1240823104672153605", "user": "DrBreaky", "date": "2020-03-20T02:10:56+00:00", "text": "Cortical source signals were computed from band-pass filtered EEG. We computed correlation coefficients between amplitude envelopes (red) of parcel signals (gray). This led to connectivity matrices for every infant for both sleep states and for each frequency band (lower panel)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThJ9_mUMAMBhhs.jpg"}]}, {"id": "1240823579647721472", "user": "DrBreaky", "date": "2020-03-20T02:12:49+00:00", "text": "3. The main effect of sleep state in the alpha band showed significant connectivity differences in two main cortical networks (red: AS > QS, blue: QS > AS).\n\nQS networks (blue) have a long typical length. AS networks have a shorter length and are predominantly occipital", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThKatLUcAA2i58.jpg"}]}, {"id": "1240824803528830976", "user": "DrBreaky", "date": "2020-03-20T02:17:41+00:00", "text": "4. Functional networks show a sleep-by-group: \n\nA positive interaction for patterns of connectivity linking occipital regions (orange)\n\nA negative interaction (cyan) for connectivity between the frontal cortex and occipital regions\n\nThese are attenuated changes in preterm infants", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThK38UUMAEHmCw.jpg"}]}, {"id": "1240825531236401153", "user": "DrBreaky", "date": "2020-03-20T02:20:34+00:00", "text": "5. This attenuation in sleep stage transitions pre-empts visual function at 2 years", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThL9uhU4AAMN_q.jpg"}]}, {"id": "1240826068761636864", "user": "DrBreaky", "date": "2020-03-20T02:22:42+00:00", "text": "6. We then recast our analyses in terms of \u201cbrain modes\u201d \u2013 a series of spatial patterns of increasing complexity that each carry a unique temporal fingerprint (waves of cortical oscillations).", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThMm82UwAAtDy6.png"}]}, {"id": "1240827184974970886", "user": "DrBreaky", "date": "2020-03-20T02:27:09+00:00", "text": "7. The experimental data (Y) are then modelled by an optimal combination of the spatial modes (m) multiplied by their temporal oscillations, f(t)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThNf5WUMAE46yy.png"}]}, {"id": "1240827693802766342", "user": "DrBreaky", "date": "2020-03-20T02:29:10+00:00", "text": "8. The model parameters for each mode are estimated from each baby brain by maximizing the match between the empirical connectivity matrices and the predicted connectivity matrices", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThOMDDUMAArz1M.png"}]}, {"id": "1240828051346210816", "user": "DrBreaky", "date": "2020-03-20T02:30:35+00:00", "text": "9. Active sleep is defined by reduced energy in a uniform mode of neural activity and increased energy in two more complex anteroposterior modes. \n\nPreterm-born infants show a deficit in this sleep-related reorganization of modal energy that carries novel prognostic information.", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThOaNoUwAAbJBF.png"}]}, {"id": "1240828633733709827", "user": "DrBreaky", "date": "2020-03-20T02:32:54+00:00", "text": "10. Brain networks (\"particles and their interactions\") and brain modes (\"waves\") are two distinct but complementary approaches of understanding large-scale brain activity.\n\nWaves provide the background dynamics context through which specific interaction can occur", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThOo9PUMAEt6jx.jpg"}]}, {"id": "1240829315660447746", "user": "DrBreaky", "date": "2020-03-20T02:35:37+00:00", "text": "Thank you very much #OHBMx !!\n\nand co-authors: @AntonTokariev @AndrewZalesky @LucaCocchi78 Xuelong Zhao, James Roberts and Sampsa Vanhatalo", "media": []}]}, {"date": "2020-03-20T02:35:12+00:00", "text": "#OHBMx-2 \u2733 #talk\n\nCsaba Orban @csabaorban*, R Kong @rubykong92, J Li @pretty_Jingwei, M W Chee @MRSleepDep, B T Yeo @bttyeo\n\n*National University of Singapore\n\n\u25b6 Time of day effects in resting state fMRI\n\n#Connectivity Sorry for the gap - seems a miscommunication with this speaker. Please check back later.", "media": [], "ids": ["1240829211222241280", "1240831365186088962"], "thread": [{"id": "1240870081757773824", "user": "csabaorban", "date": "2020-03-20T05:17:36+00:00", "text": "1. Diurnal variation is observed in many biological processes, yet time of day is rarely reported or considered in brain imaging studies. While multi-session fMRI studies often keep time of scan constant (e.g. ), this is not feasible in many large-scale studies. #OHBMX", "media": []}, {"id": "1240870644096462849", "user": "csabaorban", "date": "2020-03-20T05:19:50+00:00", "text": "2. We examined time of day effects in resting fMRI from >900 HCP subjects. We predicted higher global signal (GS) fluctuation in the afternoon (3pm) than late morning (11am), based on known circadian fluctuations in arousal and neg. corr. between arousal and GS fluctuation.", "media": []}, {"id": "1240870950389682177", "user": "csabaorban", "date": "2020-03-20T05:21:03+00:00", "text": "3. Instead we observed robust reductions in GS fluctuation throughout the day from 9am to 9pm. These effects were replicated across subjects in both sessions, and within-subjects, leveraging the fact that some subjects were scanned at different times on two days. #OHBMX", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh1dZuUcAEjCoN.jpg"}]}, {"id": "1240871620119412736", "user": "csabaorban", "date": "2020-03-20T05:23:43+00:00", "text": "4. Is this effect big enough to worry about (r=~0.2)? Perhaps you\u2019d think not, until you compare it with the sizes of typically studied behavioural-fMRI associations. Here is a comparison of time of day versus fluid intellgence effects on RSFC in the HCP (across subjects).", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh1qVGUcAIYVqa.jpg"}]}, {"id": "1240874062953345024", "user": "csabaorban", "date": "2020-03-20T05:33:25+00:00", "text": "5. Another concern is case-control studies, where patients and controls might be scanned at different times on average. I guess we won\u2019t know, until all studies start reporting time of day. #OHBMX https://t.co/1QOwHR4uG9", "media": []}, {"id": "1240875428547608576", "user": "csabaorban", "date": "2020-03-20T05:38:51+00:00", "text": "6. For a more in-depth discussion of the potential mechanisms, check out our recent paper in @PlosBiology. #Thankyou #OHBMX #StayAtHome https://t.co/YyWMhrz5RZ", "media": []}]}, {"date": "2020-03-20T02:45:53+00:00", "text": "#OHBMx-3 \u2733 #talk\n\nAshley York @AshYork1*, S Bollman @sbollmann_MRI, C Condon @UQ, M Barth @MarkusBarth2, R Cunnington @CunningtonLab, A Puckett @pucketta_m\n\n*University of Queensland\n \n\u25b6 Cortical-depth-dependent analysis of fingertip maps in human S1 using 7T fMRI\n\n#Attention #Sensory", "media": [], "ids": ["1240831902208028672", "1240832006977548288"], "thread": [{"id": "1240832183591288832", "user": "AshYork1", "date": "2020-03-20T02:47:00+00:00", "text": "1.Most cortical-depth-dependent fMRI studies focus on measuring the degree of activation at different depths. Here, we assess the consistency of the spatial distribution of responses across depth with a depth-dependent analysis of somatotopic digit maps in human S1.#OHBMx", "media": []}, {"id": "1240832537141764097", "user": "AshYork1", "date": "2020-03-20T02:48:25+00:00", "text": "2.Two types of somatotopic digit maps were obtained at sub-millimeter resolution, driven primarily by 1) bottom-up or 2) top-down processes. Maps were generated via phase-encoded vibrotactile stimulation or by sweeping attention across the fingertips.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThSevaU8AUh3Ho.jpg"}]}, {"id": "1240833391513047040", "user": "AshYork1", "date": "2020-03-20T02:51:48+00:00", "text": "3.Anatomicals were used to guide an equivolume layering approach to define a precise set of laminar surfaces, onto which functional data were interpolated and smoothed tangential to the surface. A response delay analysis was then performed for both experimental conditions.#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThTAJ4U8AAFq3F.mp4", "content-type": "video/mp4"}]}, {"id": "1240833932930641920", "user": "AshYork1", "date": "2020-03-20T02:53:57+00:00", "text": "4.Both sets of maps were found to vary across depth and were marked by a clear banded pattern at superficial layers, with bands reflecting fingertips. This pattern became less distinct with increasing depth. Clear evidence of individual variability was also seen.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThTu5QUcAM7XBE.jpg"}]}, {"id": "1240834345419431936", "user": "AshYork1", "date": "2020-03-20T02:55:36+00:00", "text": "5.The consistency of the digit representations varied across cortical depth. Superficially, in line with known properties of cortical magnification, most vertices preferred the index finger (delay values ~< 8s) . Deeper, the evidence for this known property weakened. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThUNteU4AEAUqR.jpg"}]}, {"id": "1240835204345810949", "user": "AshYork1", "date": "2020-03-20T02:59:00+00:00", "text": "6.What depth to use? Depends on your goal, but findings are in line with work in visual cortex showing the strongest responses and clearest maps superficially. However, signals near the surface are also the most spatially spread, likely obscuring some topographic detail.#OHBMx", "media": []}]}, {"date": "2020-03-20T03:00:00+00:00", "text": "#OHBMx-4 \u2733 #talk\n\nFernanda Ribeiro @NandaRibeiro93*, S Bollmann @sbollmann_MRI, A Puckett @pucketta_m\n\n*University of Queensland\n\n\u25b6 Predicting brain function from anatomy in humans using geometric deep learning \n\n#Anatomy #Applications #Methods #Modeling", "media": [], "ids": ["1240835455878172673"], "thread": [{"id": "1240835882581512193", "user": "NandaRibeiro93", "date": "2020-03-20T03:01:42+00:00", "text": "(1) The functional organization of human visual cortex is tightly coupled with the underlying anatomy. Here we developed a geometric deep learning model capable of exploiting the actual structure of the cortex to learn this complex relationship #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThVnmPU8AI7iS2.jpg"}]}, {"id": "1240836609257308160", "user": "NandaRibeiro93", "date": "2020-03-20T03:04:35+00:00", "text": "(2) The spatial organization of the retina is maintained and reflected in nearly all cortical visual areas. This retinotopic organization is similar across people, although variability exists and has been shown to be related to the underlying cortical folding pattern #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThWTDPUYAAW-Ng.jpg"}]}, {"id": "1240837120786182144", "user": "NandaRibeiro93", "date": "2020-03-20T03:06:37+00:00", "text": "(3) Retinotopic maps are often represented on cortical surface models as they only make sense considering their layout with respect to the various sulci and gyri. However, traditional CNNs do not capture the spatial information in such irregular representations #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThWurUU8AUT45w.mp4", "content-type": "video/mp4"}]}, {"id": "1240837795955892230", "user": "NandaRibeiro93", "date": "2020-03-20T03:09:18+00:00", "text": "(4) Here, we built a #geometricdeeplearning model to predict the functional organization of human visual cortex from underlying anatomical properties using the  7T dataset, which includes surface-based anatomical and functional data from 181 participants #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThXX8zUMAAla4d.jpg"}]}, {"id": "1240838325960753152", "user": "NandaRibeiro93", "date": "2020-03-20T03:11:25+00:00", "text": "(5) We demonstrate that our neural network accurately predicted the main features of the retinotopic maps. Isopolar angle and isoeccentricty contours, for example, were predicted at similar locations to those in the empirical maps #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThX3ABUYAAe-aA.mp4", "content-type": "video/mp4"}]}, {"id": "1240838930263494659", "user": "NandaRibeiro93", "date": "2020-03-20T03:13:49+00:00", "text": "(6) Additionally, we show that our neural network is able to predict nuanced variations in the retinotopic maps across individuals based solely on underlying anatomical variation. For more details, check out our preprint https://t.co/brFcDYmctz #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThYZuaVAAAD95T.mp4", "content-type": "video/mp4"}]}, {"id": "1240839991023947778", "user": "NandaRibeiro93", "date": "2020-03-20T03:18:02+00:00", "text": "Thanks to my supervisor @pucketta_m and our collaborator @sbollmann_MRI for their great support \ud83d\ude0d\ud83d\udc4f", "media": []}]}, {"date": "2020-03-20T03:15:00+00:00", "text": "#OHBMx-5 \u2733 #talk\n\nL\u00e9onie Borne @LeonieBorne*, D Rivi\u00e8re, J-F Mangin @ManginJf, M Mancip\n\n*HMRI, University of Newcastle\n\n\u25b6 Read sulcal lines through deep learning with BrainVISA\n\n#Anatomy #Methods #Tools", "media": [], "ids": ["1240839227794845696"], "thread": [{"id": "1240839624592773122", "user": "LeonieBorne", "date": "2020-03-20T03:16:34+00:00", "text": "(1) Ready to hear your #fortune? Instead of reading the palm lines, can we read the #brain lines? It may sound crazy, and yet the cortical sulci patterns reflect the functional organization of the brain. They are impacted by your past and future. But how to decipher them?\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThY5oNUcAETkY8.mp4", "content-type": "video/mp4"}]}, {"id": "1240840194619609088", "user": "LeonieBorne", "date": "2020-03-20T03:18:50+00:00", "text": "(2) It's difficult to study the cortical lines from a #MRI scan... To remedy this, the Morphologist/BrainVISA #pipeline (https://t.co/ro4m8xFzQk) allows (1) to represent the sulci in 3D and (2) to label them automatically. Until now, step (2) was making obvious errors...\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThZTM5UUAEONfP.jpg"}]}, {"id": "1240841174115438592", "user": "LeonieBorne", "date": "2020-03-20T03:22:44+00:00", "text": "(3) To improve the pipeline, we propose to use a brain-inspired model: the famous #DeepLearning (the same one that beats world Go champions and makes Siri talk)! The new model is based on the U-Net 3D neural network architecture, specific for biomedical image segmentation.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThZzSNU4AIlIC4.jpg"}]}, {"id": "1240841775855456257", "user": "LeonieBorne", "date": "2020-03-20T03:25:07+00:00", "text": "(4) UNET is better than the previous SPAM model! Yet, it requires a regularization with bottom-up geometric constraints (UNET+reg). As these constraints are sometimes excessive, we use a top-down perspective which creates a new junction when required (UNET+reg+cut).\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThalC3U4AAatUq.mp4", "content-type": "video/mp4"}]}, {"id": "1240842318741032960", "user": "LeonieBorne", "date": "2020-03-20T03:27:17+00:00", "text": "(5) UNET reproduces results that were only seen with manual labeling. E.g. with manual labeling, the central sulcus is longer in the right hemisphere than in the left (I<0) for left-handers and vice versa for right-handers. UNET finds a significant difference, not SPAM!\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThbMLBUUAAN4pW.jpg"}]}, {"id": "1240842654964797440", "user": "LeonieBorne", "date": "2020-03-20T03:28:37+00:00", "text": "(6) You now have everything in hand to search for your #heart line on your cortex\u2026\nBut in the meantime, satisfy your #love of science with #OHBMx (with a glass of wine in your sofa, it's even better). Cheers!\nMore info : https://t.co/A4U3C2B7l2", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThblmcUMAIuARw.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T03:30:00+00:00", "text": "#OHBMx-6 \u2733 #talk\n\nYu-Chi Chen @cyctbdbw*, A Fornito @AFornito, K Aquino @Kevin_M_Aquino\n\n*Monash University\n\n\u25b6 Novel asymmetry signatures for subject identification\n\n#Anatomy #Modeling #Tools", "media": [], "ids": ["1240843002731470849"], "thread": [{"id": "1240843548905226247", "user": "cyctbdbw", "date": "2020-03-20T03:32:10+00:00", "text": "1. We develop a shape asymmetry signature (SAS), which summarizes brain asymmetry across spatial scales. We show that individual\u2019s SAS is a highly unique feature that allows accurate subject identification at relatively coarse spatial scales. #OHBMx", "media": []}, {"id": "1240844017325076481", "user": "cyctbdbw", "date": "2020-03-20T03:34:02+00:00", "text": "2. We used longitudinal images from 200 healthy participants from the OASIS-3. Following the framework of ShapeDNA (Reuter et al., 2006), we used Laplace-Beltrami eigenvalues (EVs) as global geometry descriptors of the cortical surface meshes. #OHBMx", "media": []}, {"id": "1240844440870084608", "user": "cyctbdbw", "date": "2020-03-20T03:35:43+00:00", "text": "3. These EVs reflect spatial variation at discrete spatial scales. We subtracted the EVs spectra for white surface of the left hemisphere from those for the right hemisphere in the same subject at each scale, which we denote as the SAS for that subject. #OHBMx", "media": []}, {"id": "1240845285108023296", "user": "cyctbdbw", "date": "2020-03-20T03:39:04+00:00", "text": "4. We used the Euclidean distance to calculate the distances between different images both within and between subjects from the first and a later time point. We revised Amico (2015)\u2019s approach to receiving the identification score and repeated for a range of EVs. #OHBMx", "media": []}, {"id": "1240846089646817282", "user": "cyctbdbw", "date": "2020-03-20T03:42:16+00:00", "text": "5. The number to maximize the peak identification score. The SAS can serve as a unique feature for subject identification. Fig 1, the identification score across different numbers of EVs and the Euclidean distance matrix for the SAS from the T1 and T2 white surface. #OHBMx", "media": []}, {"id": "1240846675846938624", "user": "cyctbdbw", "date": "2020-03-20T03:44:35+00:00", "text": "6. The peak identifiability is given by the combination of the SAS calculated with the first 66 EVs, which represent low frequency spatial scales (Fig. 2). Surprisingly, the findings indicate that the uniqueness of asymmetry occurs at coarse spatial scales. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThfUXLU0AA3oTI.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/EThfcXpUMAA6iY2.jpg"}]}]}, {"date": "2020-03-20T03:45:00+00:00", "text": "#OHBMx-7 \u2733 #talk\n\nDavid Wright @D_K_Wright*, G Symons @GeorgiaSymmons, W O'Brien @Will_T_OBrien, T O'Brien @brien_terence, S Shultz @shultz_sandy\n\n*Monash University\n\n\u25b6 Advanced Neuroimaging as a Biomarker for Concussion\n\n#Connectivity #Methods #Neurology #Tools", "media": [], "ids": ["1240846777361879045"], "thread": [{"id": "1240848153286299651", "user": "D_K_Wright", "date": "2020-03-20T03:50:28+00:00", "text": "1. Despite increasing awareness surrounding the risks of sports-related #concussion (SRC), the diagnosis and clinical management of athletes is largely guided by subjective and self-reported symptoms which may fail to reflect whether the \ud83e\udde0 has fully recovered. #OHBMx", "media": []}, {"id": "1240849567756607488", "user": "D_K_Wright", "date": "2020-03-20T03:56:05+00:00", "text": "We need objective diagnostic and prognostic biomarkers that can better inform clinical decisions surrounding SRC. Advanced diffusion-weighted imaging (DWI) may hold the \ud83d\udd11! #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1240849466208309248/pu/vid/1280x640/bkWWzn82sMKV3npB.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1240849836439506944", "user": "D_K_Wright", "date": "2020-03-20T03:57:09+00:00", "text": "3. Using DWI, we assessed the brains of #afl athletes with SRC (n=8). Athletes were scanned at 48 hrs post-injury and again at 2 weeks post-injury (ie when they were cleared to return to play). Their scans were compared to a cohort of non-concussed afl controls (n=9). #OHBMx", "media": []}, {"id": "1240851254810562561", "user": "D_K_Wright", "date": "2020-03-20T04:02:47+00:00", "text": "4. Fixed-based analysis of fibre density revealed robust differences between afl controls and recently concussed afl athletes at 48 hrs post-injury.", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1240851203228987392/pu/vid/1280x640/JFVodgHlRxyqmSrK.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1240851926859673600", "user": "D_K_Wright", "date": "2020-03-20T04:05:27+00:00", "text": "5. Significantly, although afl athletes were asymptomatic there was still evidence of brain changes 2 weeks later. These results may indicate increased cerebral vulnerability that can persist beyond the resolution of symptoms that typically defines recovery. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1240851892470575104/pu/vid/1280x640/5Mhz1L8hIAM3oKYp.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1240852085299535877", "user": "D_K_Wright", "date": "2020-03-20T04:06:05+00:00", "text": "6. Why do we care? By objectively assessing SRC and recovery, we can potentially minimise those adverse outcomes associated with repeated injuries. Importantly, this research applies to not only SRC but mild TBI in all populations which is a nice Segway to our next tweeter #OHBMx", "media": []}]}, {"date": "2020-03-20T04:00:00+00:00", "text": "#OHBMx-8 \u2733 #talk\n\nAkram Zamani @AkiZamani*, D Wright @D_K_Wright, L Willis, L Dill, T O'Brien @brien_terence, B Semple @SciSemple\n\n*Monash University\n\n\u25b6 White matter dynamics post paediatric traumatic brain injury IN MICE\n\n#Developmental #Neurology #Social", "media": [], "ids": ["1240850552642297857"], "thread": [{"id": "1240851979112345601", "user": "AkiZamani", "date": "2020-03-20T04:05:40+00:00", "text": "Hi \ud83d\udc4b 1.#TBI is highly prevalent in children under 4. At this age the brain is incredibly plastic, with myelination and complex networks (eg those required for social interaction) not yet fully formed. As such, injury to the immature\ud83e\udde0is quite different to TBI in adults #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThkSxPUcAAa9QP.mp4", "content-type": "video/mp4"}]}, {"id": "1240852280015876096", "user": "AkiZamani", "date": "2020-03-20T04:06:52+00:00", "text": "2. A link between childhood white matter (WM) damage and later life social dysfunction has been revealed. But how does the brain change during development? And how is this effected by injury? This is what I try to answer #OHBMx", "media": []}, {"id": "1240853193828884480", "user": "AkiZamani", "date": "2020-03-20T04:10:30+00:00", "text": "3. Using an experimental pediatric \ud83d\udc2dmodel of TBI, we compared two severities of injury (mild & severe) to control mice. Longitudinal DWI was acquired at acute and chronic time points post-injury and social behavioral tests were performed at adulthood #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThkv4qUUAABajS.jpg"}]}, {"id": "1240854064939720704", "user": "AkiZamani", "date": "2020-03-20T04:13:57+00:00", "text": "4.We found acute \u2b07\ufe0fFA in regions close to injury, WM changes continued to evolve over time, extending into deeper structures and across to the contralateral hemisphere (\u2b06\ufe0fRD, \u2b07\ufe0fAD) @ adulthood,\ud83d\udc2ddeveloped dysfunction in their sociability and a tendency towards aggression #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThmLz6U8AAzk5n.jpg"}]}, {"id": "1240854456410836993", "user": "AkiZamani", "date": "2020-03-20T04:15:31+00:00", "text": "5. These results suggest that acute pathological changes in WM myelination & axonal damage precede the onset of social behavior deficits and may contribute to such behavioral manifestations #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThmi8zU0AA9g4D.mp4", "content-type": "video/mp4"}]}, {"id": "1240854782199259136", "user": "AkiZamani", "date": "2020-03-20T04:16:48+00:00", "text": "6. So what\u2019s the take home msg? \ud83d\udc76 TBI initiates a disease process that continues into adulthood. We can use this time \u231b\ufe0fto predict those children at risk of later life social dysfunction and potentially treat with therapeutic interventions #OHBMx", "media": []}]}, {"date": "2020-03-20T04:16:12+00:00", "text": "#OHBMx-9 \u2733 #talk\n\nJames M. Shine @jmacshine*, E. M\u00fcller @eli_j_muller, B. Munn @BrandonMunn11, L. Hearne, @LukeJHearne, J.B. Smith, B. Fulcher @bendfulcher, L. Cocchi @LucaCocchi78\n\n*The University of Sydney\n \n\u25b6 Core and Matrix Thalamic Sub-Populations Relate to Spatio-Temporal Cortical Connectivity Gradients\n\n#Thalamus #Timescale #Gradients", "media": [], "ids": ["1240854628327014400", "1240855036873211904"], "thread": [{"id": "1240855122139164673", "user": "jmacshine", "date": "2020-03-20T04:18:09+00:00", "text": "Much of our understanding of whole-brain functional organisation is based on the circuitry of the cerebral cortex. For instance, there is a well-known granular-to-agranular gradient that stretches from sensory to associative to limbic cortices #OHBMX (1/6)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThnJE5UUAABUsa.jpg"}]}, {"id": "1240855252078694400", "user": "jmacshine", "date": "2020-03-20T04:18:40+00:00", "text": "But the cortex isn\u2019t acting on its own\u2026 It has plenty of help from subcortical structures, like the thalamus (bonus GIF #1) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThnRGmU8AAxYlU.mp4", "content-type": "video/mp4"}]}, {"id": "1240855355304718336", "user": "jmacshine", "date": "2020-03-20T04:19:05+00:00", "text": "Zooming in: there is actually a diversity of cells within the thalamus that have distinct projection schemes: core nuclei (green) send targeted projections to granular layers of cortex, whereas matrix nuclei (blue) project more diffusely to supragranular cortex (2/6) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThnW-hUMAEqV9U.jpg"}]}, {"id": "1240855475245080576", "user": "jmacshine", "date": "2020-03-20T04:19:33+00:00", "text": "The targeted inputs form the core can be though of as driving the system in a feedforward mode, whereas the diffuse inputs from the matrix cells are more likely to shift the system into a feedback mode \u2013 kind of like raising the temperature of the cortex (bonus GIF #2) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThneJEUEAAnrVK.mp4", "content-type": "video/mp4"}]}, {"id": "1240855623543087105", "user": "jmacshine", "date": "2020-03-20T04:20:09+00:00", "text": "We used  from to identify mRNA expression of CALB1 (blue) and PVALB (green) in thalamic voxels, and then asked whether 7T rfMRI BOLD time series from these regions covaried with cortical voxel time series (N = 60). L: thalamus. R: cortical correlations (3/6) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThnmu_U8AEnWYv.jpg"}]}, {"id": "1240855719831670785", "user": "jmacshine", "date": "2020-03-20T04:20:32+00:00", "text": "We had found multimodal evidence that the thalamus and cortex were bound together in large-scale spatial gradients (bonus GIF #3) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThnsYCUUAAUKEu.mp4", "content-type": "video/mp4"}]}, {"id": "1240855785657094145", "user": "jmacshine", "date": "2020-03-20T04:20:47+00:00", "text": "The cortical patterns were strongly positively correlated with the first gradient of vertex-wise functional connectivity calculated from a different dataset by Margulies et al., 2016 (PNAS) (4/6) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThnwSiUEAAPVza.jpg"}]}, {"id": "1240855882621046784", "user": "jmacshine", "date": "2020-03-20T04:21:11+00:00", "text": "The thalamus helps the brain ride cortical gradients (bonus GIF #4) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThn109UcAAbUVQ.mp4", "content-type": "video/mp4"}]}, {"id": "1240855987180859395", "user": "jmacshine", "date": "2020-03-20T04:21:35+00:00", "text": "@BrandonMunn11 calculated the Hurst exponent of each cortical region, which is an estimate of the intrinsic time-scale. We observed a positive correlation that suggests that the cortical regions supported by Matrix thalamus have a longer intrinsic time-scale #OHBMx (5/6)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThn8AFUYAMDTHd.jpg"}]}, {"id": "1240856099772715008", "user": "jmacshine", "date": "2020-03-20T04:22:02+00:00", "text": "The diffuse inputs from the thalamic matrix cells to the cortex are helping the system to function over longer time-scales (bonus GIF #5) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThoCfnU0AAG835.mp4", "content-type": "video/mp4"}]}, {"id": "1240856160296529921", "user": "jmacshine", "date": "2020-03-20T04:22:17+00:00", "text": "And @EliMuller found a correlation between cortical dynamic FC variability and the cortical regions supported by diffuse matrix inputs. These regions had both longer time-scale and greater variability. How cool is the thalamus!?! https://t.co/V0ZXJMonJ1 (6/6) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThoGA8U8AICiUo.png"}]}, {"id": "1240856279452532741", "user": "jmacshine", "date": "2020-03-20T04:22:45+00:00", "text": "Many thanks from my co-authors @EliMuller @BrandonMunn11 @LucaCocchi78 @LukeJHearne @bendfulcher and Jared Smith. Lucky to have such a great team to work on the project! (bonus GIF #6) #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/EThoM80UcAIvpeb.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T04:30:12+00:00", "text": "#OHBMx-10 \u2733 #talk\n\nSidhant Chopra @SidChop*, A Fornito @AFornito, S M . Francey @FranceyShona, B O\u2019Donoghue @BrianNODonoghue, V Cropley @CropleyVanessa, B Nelson, J Graham, L Baldwin, S Tahtalia, H P Yuen, K Allott @kelly_allott, M Alvarez-Jimenez @MarioAlvarezJi1,   S Harrigan, K Sabaroedin , C Pantelis , S J Wood , P McGorry \n\n*Turner Institute of Brain and Mental Health\n\n\u25b6 Brain Volume Reduction in Psychosis: Results from a Placebo-controlled RCT\n\n#Anatomy #Psychiatry", "media": [], "ids": ["1240858153291354114", "1240859249766354946"], "thread": [{"id": "1240859630059663360", "user": "SidChop", "date": "2020-03-20T04:36:04+00:00", "text": "[1] Thousands of studies (!!) report \ud83e\udde0 volume changes in psychosis. Some show these changes can progress with illness duration. A major unresolved question is whether these changes are caused by underlying 'illness' or the \ud83d\udc8a antipsychotics themselves #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThrPiaUcAIOuYE.jpg"}]}, {"id": "1240860344609718274", "user": "SidChop", "date": "2020-03-20T04:38:54+00:00", "text": "[2] The best way to address this question is through a placebo-controlled trial in patients who have not yet been exposed to \ud83d\udc8a . The team at @Orygen_aus recently overcame practical and ethical challenges associated with such a trial. Checkout -> https://t.co/RIyRPB56c5\n#OHBMx", "media": []}, {"id": "1240860855333351424", "user": "SidChop", "date": "2020-03-20T04:40:56+00:00", "text": "[3] In a triple-blind \ud83d\ude11\ud83d\ude11\ud83d\ude11 randomisation patients got psychosocial therapy plus an atypical antipsychotic or a placebo over 6 months. A healthy control group also scanned. MRI at 3 time-points. Method: Longitudinal mixed-effects VBM + wild-bootstrapping for inference. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThsNfQU8AAiUD6.jpg"}]}, {"id": "1240861537507532801", "user": "SidChop", "date": "2020-03-20T04:43:39+00:00", "text": "[4] Primary finding: within the pallidum (A) volume \u2197\ufe0fincreased in the antipsychotic group, \u2198\ufe0fdecreased in the placebo group and \u27a1\ufe0fstable in healthy control (B). Greater increase was associated with REDUCED symptoms (C). This looks like \ud83d\udc8a-induced neuroprotection. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThs-mJUcAA_XEG.jpg"}]}, {"id": "1240861828277653507", "user": "SidChop", "date": "2020-03-20T04:44:48+00:00", "text": "[5] Secondary findings: Some preliminary evidence that frontal cortex volume loss is due to underlying illness and unaffected by \ud83d\udc8a. We also found possible neurotoxicity associated with \ud83d\udc8a within the cerebellum - but these two findings need further investigation \ud83e\uddd0. \n#OHBMx", "media": []}, {"id": "1240862335633256449", "user": "SidChop", "date": "2020-03-20T04:46:49+00:00", "text": "[6] Overall, looks like experiencing psychosis and antipsychotic exposure have different effects on \ud83e\udde0 volume. The most robust effect is consistent with a neuroprotective effect of atypical antipsychotics on pallidum volume ..at least in the early stages of treatment\u2695\ufe0f.\n#OHBMx", "media": []}]}, {"date": "2020-03-20T04:45:46+00:00", "text": "#OHBMx-11 \u2733 #talk\n\nKristina Sabaroedin @kristinasabr*, A Razi @adeelrazi, K Aquino @Kevin_M_Aquino, S Chopra @SidChop, B Nelson, K Allott @kelly_allot, M Alvarez-Jimenez @MarioAlvarezJi1, J Graham, L Baldwin, S Tahtalian, H P Yuen, S Harrigan, V Cropley @CropleyVanessa,   C Pantelis , S Wood , B O\u2019Donoghue, S Francey , P McGorry , A Fornito \n\n*Monash University\n\n\u25b6 Effective connectivity of frontostriatal systems in first-episode psychosis\n\n#Connectivity #Modeling #Psychiatry", "media": [], "ids": ["1240862070041505793", "1240862269149335552"], "thread": [{"id": "1240862648712876032", "user": "kristinasabr", "date": "2020-03-20T04:48:04+00:00", "text": "1. Connectivity of frontostriatal circuits is altered in the psychosis spectrum. We used dynamic causal modelling (DCM) to examine whether dysconnectivity stems from disrupted bottom-up or top-down signaling of frontostriatal systems in first-episode psychosis (FEP) #OHBMx", "media": []}, {"id": "1240863097323057153", "user": "kristinasabr", "date": "2020-03-20T04:49:51+00:00", "text": "2. 52 FEP patients and 22 healthy controls (HCs) underwent resting-state fMRI. Biologically plausible connections between eight regions in the left hemisphere encompassing the frontostriatal systems were modelled using spectral DCM. #OHBMx", "media": []}, {"id": "1240863548009353216", "user": "kristinasabr", "date": "2020-03-20T04:51:38+00:00", "text": "3. The regions are: DLPFC, VMPFC, anterior hippocampus, amygdala, dorsal caudate, nucleus accumbens, thalamus, and the midbrain. Differences in effective connectivity (EC) between patients and HCs was assessed using a parametric Bayesian model. #OHBMx", "media": []}, {"id": "1240864214169694214", "user": "kristinasabr", "date": "2020-03-20T04:54:17+00:00", "text": "In FEPs vs HCs: increased bottom-up EC from midbrain to anterior hippocampus and to nucleus accumbens, reduced EC from thalamus to midbrain, increased inhibitory self-connections of DLPFC and midbrain, and reduced self-connection of VMPFC #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThvTpIUUAAZ5C7.jpg"}]}, {"id": "1240864839452332032", "user": "kristinasabr", "date": "2020-03-20T04:56:46+00:00", "text": "5. Increased ECs from DLPFC, thalamus, and anterior hippocampus to the midbrain as well as reduced effective connections from midbrain to thalamus and from nucleus accumbens to hippocampus were associated with positive symptoms in FEPs. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThvx7DVAAAe3j0.jpg"}]}, {"id": "1240865283167772672", "user": "kristinasabr", "date": "2020-03-20T04:58:32+00:00", "text": "6. Our results are consistent with animal models of psychosis (https://t.co/5y0iCBDUFq); esp increased ECs from the midbrain. Associations between positive symptoms and EC parameters are also in line with elevated dopamine activity in psychosis. Thanks for tuning in! #OHBMx", "media": []}]}, {"date": "2020-03-20T05:00:57+00:00", "text": "#OHBMx-12 \u2733 #keynote\n\nThomas Yeo @bttyeo*\n\n*National University of Singapore\n\n\u25b6 KEYNOTE: Incorporating gradients in large-scale biophysically-plausible circuit models https://t.co/EHt5LukYKn", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThw8PqU4AAjmu8.png"}], "ids": ["1240865893111828480"], "thread": [{"id": "1240866208032763907", "user": "bttyeo", "date": "2020-03-20T05:02:12+00:00", "text": "1. First, I hope everyone & their families are safe from the coronavirus! I also like to thank my lab & collaborators, who did the actual work:  Xiaolu Kong    Gustavo Deco,  \n\nWith that, let's start! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThxNTRVAAAcZua.jpg"}]}, {"id": "1240866360764157952", "user": "bttyeo", "date": "2020-03-20T05:02:49+00:00", "text": "2. A powerful approach to bridge microscale and macroscale brain organization is the use of large-scale biophysically-plausible circuit models of coupled brain regions. See work from       and many others. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThxXDGUYAA1vuA.jpg"}]}, {"id": "1240866541886795777", "user": "bttyeo", "date": "2020-03-20T05:03:32+00:00", "text": "3. These large-scale circuit models are powered by biophysically plausible local neural mass models (NMMs) mathematically linked by anatomical connections. The resulting large-scale circuit models can generate realistic fMRI with relatively low parametric complexity. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThxiTNUMAAyP8a.jpg"}]}, {"id": "1240866629308657665", "user": "bttyeo", "date": "2020-03-20T05:03:53+00:00", "text": "4. Most large-scale circuit studies assume that local circuit properties are uniform across the brain, which is not biologically plausible. We (https://t.co/b6GRDrnYAz) relax this assumption & automatically estimate the parameters based on fit to static functional connectivity.", "media": []}, {"id": "1240866739065253889", "user": "bttyeo", "date": "2020-03-20T05:04:19+00:00", "text": "5. The fit was a lot better in unseen test data. Without assuming the existence of a hierarchy, the estimated micro-circuit parameters revealed a large-scale gradient, matching well with a meta-analysis, T1/T2 myelin estimate, resting-fMRI principal gradient & histology. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThxtLOUwAA5c3L.jpg"}]}, {"id": "1240866811844808707", "user": "bttyeo", "date": "2020-03-20T05:04:36+00:00", "text": "6. @johndmurray has a similar paper (https://t.co/Grbl6hrqQr) around the same time, but recurrent strength was in opposite direction to ours, suggesting degeneracies (not good!!). Are degeneracies inherent in neural mass models? Or maybe our estimation algorithms suck? Or both?", "media": []}, {"id": "1240867041315188736", "user": "bttyeo", "date": "2020-03-20T05:05:31+00:00", "text": "7. Our approach (Wang2019) also fitted time-varying FC poorly. So, we extend @johndmurray approach to fit both static & time-varying FC. We achieved much better fit with time-varying FC in leave-out test data. Surprisingly, fit was also better for static FC! Is that free lunch?", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThx_LVU8AAITDI.jpg"}]}, {"id": "1240867139508035585", "user": "bttyeo", "date": "2020-03-20T05:05:54+00:00", "text": "8. More free lunch: across many initializations & 2 parcellations, gradient directions were highly consistent. Recurrent strength (@johndmurray is right!) & noise amplitude increase from sensory-motor to association cortex, while external input goes in opposite direction. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThyEkzUwAEMDpz.jpg"}]}, {"id": "1240867262124285953", "user": "bttyeo", "date": "2020-03-20T05:06:24+00:00", "text": "9. An open question is whether sharp transitions in time-varying FC are due to switching of discrete brain states. We find that sharp transitions in sliding-window FC correspond to sliding-window standard deviation of regional BOLD signals in both real and simulated data. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThyLBkUMAIlyV5.jpg"}]}, {"id": "1240867344898899968", "user": "bttyeo", "date": "2020-03-20T05:06:43+00:00", "text": "10. Transitions in sliding-window (sw) STD in sensory-motor regions appear to drive sharp transitions in sw-FC in real & simulated data. Thus, biophysically plausible mean field models can generate realistic time-varying FC without explicit discrete brain states. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/EThyQbPUYAAzuKg.jpg"}]}, {"id": "1240867496816603137", "user": "bttyeo", "date": "2020-03-20T05:07:20+00:00", "text": "Thanks for \u201clistening\u201d. And thanks to the organizers @OHBMequinoX   for inviting me #OHBMx", "media": []}]}, {"date": "2020-03-20T05:30:16+00:00", "text": "#OHBMx-13 \u2733 #talk\n\nAlexander Puckett @pucketta_m*, M Schira, Z Isherwood @zoeyisherwood, J Victor, J Roberts, M Breakspear @DrBreaky\n\n*University of Queensland\n\n\u25b6 Manipulating the structure of natural images using wavelets to probe the visual hierarchy\n\n#Methods #Sensory", "media": [], "ids": ["1240873270439579648"], "thread": [{"id": "1240873591777792000", "user": "pucketta_m", "date": "2020-03-20T05:31:33+00:00", "text": "(1) There is mounting evidence suggesting that the cortex is strongly \u2018tuned\u2019 to the statistical properties of naturalistic stimuli. Here we provide a technique to manipulate these naturalistic stimuli with tight experimental control. #OHBMx", "media": []}, {"id": "1240873817632694273", "user": "pucketta_m", "date": "2020-03-20T05:32:27+00:00", "text": "(2) All natural images, regardless of specific content (e.g., forests, a patch of pebbles, etc.), share low- and high-order statistical properties. Using wavelets, we show how the higher-order properties can be manipulated while keeping lower-level properties intact. #OHBMx", "media": []}, {"id": "1240874243169996801", "user": "pucketta_m", "date": "2020-03-20T05:34:08+00:00", "text": "(3) Structure can be disrupted at specific spatial scales: (a) intact image, (b) fine scale degraded, (c) coarse scale degraded, (d) all scales except fine scale degraded, and (e) all scales degraded. Lower-level properties such as spatial frequency are preserved (f). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh4RLjU0AI9uFB.jpg"}]}, {"id": "1240875000598388741", "user": "pucketta_m", "date": "2020-03-20T05:37:09+00:00", "text": "(4) Using our \u2018wavestrapping\u2019 technique it is also possible to (a) target a specific region rather than the entire image, (b) disrupt the structure in color images, and (c) extend the approach to dynamic (e.g., film) stimuli. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh46ifUwAEytCr.jpg"}]}, {"id": "1240875484247773184", "user": "pucketta_m", "date": "2020-03-20T05:39:04+00:00", "text": "(5) Using fMRI, we then showed that \u2013 as expected \u2013 responses at different levels of the visual hierarchy (a) are differentially sensitive to the wavelet manipulations (b). Note: N1 is structured, N2 has all structure degraded, and N3 has one scale of structure present. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh5evFU8AEPpNW.jpg"}]}, {"id": "1240875898263318528", "user": "pucketta_m", "date": "2020-03-20T05:40:43+00:00", "text": "(6) In summary, our work shows how to use wavelets to parametrically and subtly manipulate the complex statistical properties of natural scenes with a high degree of control and flexibility \u2013 and that the visual system is sensitive to these subtle manipulations. #OHBMx", "media": []}, {"id": "1240876149493784576", "user": "pucketta_m", "date": "2020-03-20T05:41:43+00:00", "text": "Many thanks to the conference organizers as well as the co-authors (@drBreaky, @zoeyisherwood, Mark Schira, Jonathan Victor, and James Roberts)!", "media": []}]}, {"date": "2020-03-20T05:45:03+00:00", "text": "#OHBMx-14 \u2733 #talk\n\nTom Johnstone @itjohnstone*\n\n*Swinburne University of Technology\n\n\u25b6 Are you local? In brain imaging, computer says \u201cno\u201d\n\n#Methods #Tools", "media": [], "ids": ["1240876990678183946"], "thread": [{"id": "1240877496771309568", "user": "itjohnstone", "date": "2020-03-20T05:47:04+00:00", "text": "1. Anatomical precision important in fMRI (Devlin & Poldrack, 2007). Extensive efforts to improve resolution, registration, parcellation. Thousands of manuscripts make claims of functionally localised activation. Almost none actually test functional localization.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh7dx0UwAApj0u.png"}]}, {"id": "1240878277696188416", "user": "itjohnstone", "date": "2020-03-20T05:50:10+00:00", "text": "2. Voxelwise thresholding of statistical map, with locations of suprathreshold activation used as a proxy for localization. Approach implicitly accepts null hypothesis  areas falling below threshold are not activated. Otherwise would not claim \"localisation\" in abstract! #OHBMx", "media": []}, {"id": "1240878779146178561", "user": "itjohnstone", "date": "2020-03-20T05:52:10+00:00", "text": "3. Most fMRI studies under-powered (Thirion et al., 2007), so lots of false negatives. Implication: thresholded brain activation **more focal**! Studies with larger N => less focal regions of activation (Gonzalez-Castillo et al., 2012; Thyreau et al., 2012) #OHBMx", "media": []}, {"id": "1240879327308156933", "user": "itjohnstone", "date": "2020-03-20T05:54:20+00:00", "text": "4. Solutions: 1) Voxelwise Inferiority Maps: identify regions significantly less activated than activated clusters, using permutation tests 2) Spatial Mixture Models: Model fMRI as distributions of activated, non-activated voxels. Assign probabilities of membership of each #OHBMx", "media": []}, {"id": "1240880148892635144", "user": "itjohnstone", "date": "2020-03-20T05:57:36+00:00", "text": "5. 3) Bayes Factors: Calculate voxelwise Bayes Factor for Null versus activation model. Accept voxels as non-activated only when BF(null) sufficiently high. Moderate claims of localisation accordingly! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh96MWU8AAgleU.jpg"}]}, {"id": "1240880714293194752", "user": "itjohnstone", "date": "2020-03-20T05:59:51+00:00", "text": "6. Note: not claiming \u201cno activity\u201d, but placing limit on claims of localisation & identifying *plausibly* non/less active regions. All simple to implement at group analysis level (BF Python code avail from me).\nThank you #OHBMx organisers. Look after yourselves and each other!", "media": []}]}, {"date": "2020-03-20T06:00:16+00:00", "text": "#OHBMx-15 \u2733 #talk\n\nKshipra Gurunandan @KGurunandan*, J Arnaez-Telleria, M Carreiras, P M Paz-Alonso\n\n*Basque Center On Cognition, Brain and Language\n\n\u25b6 Lateralisation & plasticity: what can comparative approaches tell us?\n\n#Language #Methods", "media": [], "ids": ["1240880821340258306"], "thread": [{"id": "1240881358030770176", "user": "KGurunandan", "date": "2020-03-20T06:02:24+00:00", "text": "1 \"Language is left-lateralised\"\n\nStudies find a consistent fronto-temporo-parietal language network, but its lateralisation isn't so clear-cut\n\nMOST studies find left lateralisation, so what's up with the many that don't?\n\nLet's examine lateralisation & its plasticity #OHMBx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh-4Y8U4AA1Xkk.jpg"}]}, {"id": "1240881745567731717", "user": "KGurunandan", "date": "2020-03-20T06:03:57+00:00", "text": "2 We conducted fMRI studies with language learners:\ni intermediate vs advanced Basque learners\nii intermediate English learners pre and post intensive language classes\n\nWe looked at lateralisation of the classical language network and how this changed with language learning #OHBM", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh_XRtVAAEEGO9.jpg"}]}, {"id": "1240882219570188288", "user": "KGurunandan", "date": "2020-03-20T06:05:50+00:00", "text": "3 Production is left-lateralised, but comprehension is not!\n\nAt the group-level, comprehension could be left/bi/right-lateralised depending on the sample, cut-off for bilaterality, whether one uses mean or median\n\nBecomes clearer why different studies get different answers #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETh_vloU8AATmDE.jpg"}]}, {"id": "1240882763764346880", "user": "KGurunandan", "date": "2020-03-20T06:08:00+00:00", "text": "4 \nQ: How similar is lateralisation in the native language acquired from birth and a new language acquired later?\nA: Pretty similar when you're not very good at your new language. Less similar as you become better at your new language. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiAOpOU8AA_G_5.jpg"}]}, {"id": "1240883293005828097", "user": "KGurunandan", "date": "2020-03-20T06:10:06+00:00", "text": "5 Forget about left-right. Are languages at least in the same hemisphere? \n\nIncreasing proficiency -> languages lateralise to opposite hemispheres\n\nLongitudinal study shows subjects with languages in opposite hemispheres stay that way, while the rest switch with learning #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiAwGYUMAAJX0R.jpg"}]}, {"id": "1240883891277156352", "user": "KGurunandan", "date": "2020-03-20T06:12:28+00:00", "text": "6 Left hemisphere is specialised for production, but comprehension is much more flexible/plastic\n\u00b0 May explain psycholinguistic comprehension-production asymmetry\n\u00b0 Echoes recovery patterns in stroke patients, important to test each language modality in clinical settings #OHBMx", "media": []}]}, {"date": "2020-03-20T06:15:00+00:00", "text": "#OHBMx-16 \u2733 #talk\n\nLingbin Bian @LBIAN5*, T Cui, A Razi @adeelrazi, J Keith @ThatNinjaHoke\n\n*Monash University\n\n\u25b6 Bayesian network change point detection using weighted stochastic block model\n\n#Connectivity #Dynamics #Modeling #Networks", "media": [], "ids": ["1240884526143819776"], "thread": [{"id": "1240884979296432133", "user": "LBIAN5", "date": "2020-03-20T06:16:48+00:00", "text": "1. In dynamic brain network analysis, change point detection problem focuses to estimate the transition of the brain states without the prior knowledge of experimental design. Some schemes were proposed using spectral clustering and dynamic connectivity regression. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiCRUlUUAA1tAz.jpg"}]}, {"id": "1240885424706347009", "user": "LBIAN5", "date": "2020-03-20T06:18:34+00:00", "text": "2. We present a novel Bayesian method for detecting assignment of nodes to the communities in functional brain networks via model fitness assessment. We identify the dynamic community change points based on overlapped sliding window applied to multivariate BOLD signals. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiCoqCUEAApqa9.jpg"}]}, {"id": "1240885915746107396", "user": "LBIAN5", "date": "2020-03-20T06:20:31+00:00", "text": "3. The parameters for the model include latent label vectors that assign network nodes to the interacting communities, and the weighted stochastic block model parameter determines the weighted connectivity within and between communities. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiDCBFUcAAlnHP.jpg"}]}, {"id": "1240886421545553923", "user": "LBIAN5", "date": "2020-03-20T06:22:32+00:00", "text": "4. We use weighted stochastic block model to quantify the likelihood of a network configuration and evaluate the goodness of fit between model and observations by posterior predictive discrepancy within the sliding window for inferring the change points. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiDcPNU0AAHJVt.jpg"}]}, {"id": "1240887006420336640", "user": "LBIAN5", "date": "2020-03-20T06:24:51+00:00", "text": "5. We use Markov chain Monte Carlo to efficiently sample parameters from posterior distribution. The method is applied to synthetic data and working memory task fMRI data. The discrepancy index (DI) increases dramatically when the true change point is within the window. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiEFXgUMAEY2bk.jpg"}]}, {"id": "1240887299975426049", "user": "LBIAN5", "date": "2020-03-20T06:26:01+00:00", "text": "6. Conclusion: The fully Bayesian framework is flexible for characterising the latent properties of networks. Posterior predictive discrepancy can consistently and robustly reflect the dynamic characteristics of communities.  #OHBMx", "media": []}]}, {"date": "2020-03-20T06:30:03+00:00", "text": "#OHBMx-17 \u2733 #talk\n\nTom Schonberg @tschonberg*, R Botvinik-Nezer @rotembot, F Holzmeister @flxhlzmstr, C F Camerer @CFCamerer, A Dreber, J Huber, M Johannesson, M Kirchler, T E Nichols @ten_photos University of Oxford, R A Poldrack @russpoldrack\n\n*Tel Aviv University\n\n \u25b6 Variability of fMRI results across analysis teams and over optimism in prediction markets\n\n#Decision #Methods #Modeling\n\n", "media": [], "ids": ["1240888314137202689", "1240888369007063040"], "thread": [{"id": "1240888588318932995", "user": "tschonberg", "date": "2020-03-20T06:31:08+00:00", "text": "1 Ever wondered what would happen if 70 different groups analyzed the same fMRI dataset? @rotembot  @ten_photos @russpoldrack  https://t.co/XYidxwXdKy  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiFj5fXsAAo_Jp.jpg"}]}, {"id": "1240889004792430592", "user": "tschonberg", "date": "2020-03-20T06:32:48+00:00", "text": "2 197 neuroimagers and economists teamed up for this project! An fMRI dataset was distributed to 70 analysis teams. They analyzed the data as they usually do in their labs and tested 9 pre-defined hypotheses. e.g. Positive effect of gain in vmPFC  #OBHMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiF7ThWAAAw8HN.jpg"}]}, {"id": "1240889373136207874", "user": "tschonberg", "date": "2020-03-20T06:34:15+00:00", "text": "3 Fraction of teams reporting a significant binary result varied substantially, most hypotheses between 20-40% endorsement across teams. Prediction markets showed over-optimism about reproducibility of results, even if researchers analyzed the data themselves! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiGOCZXgAAybeL.jpg"}]}, {"id": "1240889798564470784", "user": "tschonberg", "date": "2020-03-20T06:35:57+00:00", "text": "4 Interestingly, underlying statistical brain maps were more correlated than expected based on the diverse reported results , but even highly correlated teams nonetheless reported very different corrected results #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiGhGlXkAI3MRn.jpg"}]}, {"id": "1240890213687320578", "user": "tschonberg", "date": "2020-03-20T06:37:36+00:00", "text": "5 We found that estimated smoothness of images and several other factors were related to the results. However, we don\u2019t believe there is a \u201cgolden\u201d pipeline. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiG9daXYAACWdF.jpg"}]}, {"id": "1240890560673730560", "user": "tschonberg", "date": "2020-03-20T06:38:58+00:00", "text": "6 Take home: share uncorrected data. Be aware of analytic flexibility in every scientific field with high-dim data and complex analytic choices. We\u2019re working on multiverse tools for individual studies . \nRead here https://t.co/XYidxwXdKy  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiHUUoXYAEzz63.jpg"}]}, {"id": "1240891565792538624", "user": "tschonberg", "date": "2020-03-20T06:42:58+00:00", "text": "totally! Share everything - all task, analysis, pipeline. All the codes and results are reproducible. @russpoldrack will give a talk here later tonight here @OHBMequinoX", "media": []}, {"id": "1240891944076824576", "user": "tschonberg", "date": "2020-03-20T06:44:28+00:00", "text": "Another huge thank you to all analysis teams and prediction market traders - we couldn't have done it without you! Our field has always pushed ahead for better methods and reporting. Our results apply to every field. Thanks again to @OHBMequinoX and stay safe everyone", "media": []}]}, {"date": "2020-03-20T06:45:28+00:00", "text": "#OHBMx-18 \u2733 #talk\n\nMansoureh Fahimi Hnazaee @FahimiMansoureh*, B Wittevrongel @benjaminwtv, E Khachatryan, A Libert, E Carrette @CarretteEvelien, I Dauwe, A Meurs, P Boon, D Van Roost, M M Van Hulle\n\n*KU Leuven\n \n\u25b6 Localization of deep brain activity with scalp and subdural EEG\n\n#Modeling", "media": [], "ids": ["1240892195655180288", "1240892403143168000"], "thread": [{"id": "1240892613592518656", "user": "FahimiMansoureh", "date": "2020-03-20T06:47:08+00:00", "text": "1/ Deep and subcortical regions play an important role in brain function. However, as joint recordings at multiple spatial scales in humans are scarce, it is unknown how ECoG and EEG differ in their ability to locate sources of deep brain activity. #uzgent #compNeuroKUL  #OHBMx", "media": []}, {"id": "1240893101381767169", "user": "FahimiMansoureh", "date": "2020-03-20T06:49:04+00:00", "text": "2/ ECoG is superior in locating proximal activity (no cranium, no spatial blurring). But spatial coverage is limited compared to whole scalp coverage of EEG (affects localization of distal sources). EEG studies have located activity from deep sources such as hippocampus #OHBMx", "media": []}, {"id": "1240893428680048642", "user": "FahimiMansoureh", "date": "2020-03-20T06:50:22+00:00", "text": "3/ We recorded simultaneous ECoG, EEG and depth electrode (DE) in 4 patients with refractory epilepsy during quiet wakefulness (focus on nonepileptic data). DE were in hippocampus or insula, provided us with our ground truth for source localization of deep brain activity #OHBMX", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiJ_BpWoAAAtCb.jpg"}]}, {"id": "1240893648335757312", "user": "FahimiMansoureh", "date": "2020-03-20T06:51:15+00:00", "text": "4/ Weak correlation was found for DE/ECoG and DE/EEG. Dipole modeling of ECoG and EEG ICs showed ICs correlating with DE are located closer to DE than non-correlating counterparts. Finally, source localization (SL) accuracy was compared between correlating ECoG and EEG ICs #OHBMX", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiKKp6X0AEA8Wu.jpg"}]}, {"id": "1240894087248654336", "user": "FahimiMansoureh", "date": "2020-03-20T06:52:59+00:00", "text": "5/ ECoG SL was better, but not by that much. Given unfavorable conditions for EEG (ECoG is placed close to DE, ECoG itself attenuates EEG, EEG had only 27 electrodes) we don\u2019t believe increased ECoG SL accuracy of DEEP sources large enough to justify a preference over EEG. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiKe1NWsAAr0HH.jpg"}]}, {"id": "1240894550538964993", "user": "FahimiMansoureh", "date": "2020-03-20T06:54:50+00:00", "text": "6/ More simultaneous recording can bridge spatial scales and confirm our results, as correlation could be result of common conductor (we argue against this in paper, under review). Experimental designs are needed with prior knowledge of neural generator in deep structures. #OHBMx", "media": []}, {"id": "1240895392453140480", "user": "FahimiMansoureh", "date": "2020-03-20T06:58:10+00:00", "text": "Finally, I would like to thank everyone in the team for their hard work. And despite the difficult times, happy Persian new year to all", "media": []}]}, {"date": "2020-03-20T07:00:23+00:00", "text": "#OHBMx-19 \u2733 #talk\n\nKhansa Rasheed @KhansaRasheed*, A Qayyum @adnan_qayyum020, J Qadir @junaidq, S Sivathamboo @shobi__s, P Kwan, L Kuhlmann @levink2, T O'Brien, A Razi @adeelrazi\n\n*Information Technology University\n \n\u25b6 Machine Learning for Predicting Epileptic SeizuresUsing EEG Signals: A Review\n\n#Disorders", "media": [], "ids": ["1240895946570883073", "1240896024241037318"], "thread": [{"id": "1240896218307407872", "user": "KhansaRasheed", "date": "2020-03-20T07:01:27+00:00", "text": "1) Epilepsy_a chronic brain disease. According to WHO 70 million people are suffering from this disease. 30% of the patients are resistant to medical treatments and living in the threat of disease attack anywhere anytime. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiMgUsXYAEby-l.jpg"}]}, {"id": "1240896635439325184", "user": "KhansaRasheed", "date": "2020-03-20T07:03:07+00:00", "text": "2) The following animation illustrates the history of epilepsy prediction. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETiM327X0AIEMdE.mp4", "content-type": "video/mp4"}]}, {"id": "1240897043331256320", "user": "KhansaRasheed", "date": "2020-03-20T07:04:44+00:00", "text": "3) Epilepsy seizure prediction using ML requires handcrafted features. ML algos submitted in 2016 competition were complicated.  DL algos are being used to automatically learn proper features. Multi-modal data collection is underway for efficient seizure prediction. #OHBMx.", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETiNLgmXsAInnFh.mp4", "content-type": "video/mp4"}]}, {"id": "1240897447913754625", "user": "KhansaRasheed", "date": "2020-03-20T07:06:20+00:00", "text": "4) Potential pitfalls:\n1.Unavailability of open access EEG datasets.\n2.Incapability to predict seizures using raw EEG.\n3.Rise in time & computational cost due to excessive feature extraction.\n4.DL algos too require abundant data.\n5.Lack of efficient hardware applications. #OHBMx", "media": []}, {"id": "1240897772028559361", "user": "KhansaRasheed", "date": "2020-03-20T07:07:38+00:00", "text": "5) Future Directions\n1. Good quality signal selection for effective seizure prediction.\n2. GANs can curb unavailability of long-term EEG signals w/ annotations. \n3. Making ML decisions interpretable to ease uncertainty regarding quantification of predictions. #OHBMx", "media": []}, {"id": "1240898133841846272", "user": "KhansaRasheed", "date": "2020-03-20T07:09:04+00:00", "text": "6) Conclusion\nWe highlighted the need for early seizure prediction, and how ML/DL techniques predict them.  Still, majority surveys focused only on EEG and a few covered developments of prediction techniques. We also identified future directions and open research problems. #OHBMx", "media": []}, {"id": "1240898648730472448", "user": "KhansaRasheed", "date": "2020-03-20T07:11:07+00:00", "text": "I would like to thank my all team mates due to which this was possible and to @OHBMequinoX for giving me the opportunity to present my paper.", "media": []}]}, {"date": "2020-03-20T07:15:33+00:00", "text": "#OHBMx-20 \u2733 #talk\n\nTimo Roine @Connect2Brain*, J O Nieminen, A E Tervo, L Marzetti @MarzettiLaura, V Pizzella @vittomeg, C Zrenner, R J Ilmoniemi @rjirjirji, G L Romani, U Ziemann\n\n*Aalto University\n \n\u25b6 Connecting to the networks of the human brain with multi-locus TMS\n\n#Connectivity #Decision #Instrumentation #Networks", "media": [], "ids": ["1240899767044935681", "1240899907151466496"], "thread": [{"id": "1240900245934006272", "user": "Connect2Brain", "date": "2020-03-20T07:17:28+00:00", "text": "1\n\nBrain disorders cause suffering and yearly costs of \u20ac1000bn in Europe alone. Non-invasive neuromodulation with transcranial magnetic stimulation (TMS) has demonstrated moderate to good results and has been approved widely for clinical use.\n\nhttps://t.co/DDoJgxPTPE #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiQKIxXkAA7jCp.jpg"}]}, {"id": "1240900651288272897", "user": "Connect2Brain", "date": "2020-03-20T07:19:04+00:00", "text": "2\nControlling the brain is difficult via a single site, possibly contributing to the small effect size of the single-coil therapies. Multi-locus TMS (mTMS) allows real-time control over the locus, direction, intensity, and timing of the stimulation.\nhttps://t.co/Jva4HvAyPO #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiQiA9WsAAJxVF.jpg"}]}, {"id": "1240901018847711232", "user": "Connect2Brain", "date": "2020-03-20T07:20:32+00:00", "text": "3\nThe goals in ConnectToBrain project:\n1) mTMS device covering most of the cortex\n2) feedback-controlled stimulation based on brain activity and connectivity\n3) demonstration of safety and therapeutic utility in Alzheimer\u2019s disease and motor stroke\nhttps://t.co/EHONWHkUBj #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiQ3FKX0AACuMx.jpg"}]}, {"id": "1240901472516222977", "user": "Connect2Brain", "date": "2020-03-20T07:22:20+00:00", "text": "4\nMulti-locus TMS enables effortless automation of, e.g., target search. We developed an algorithm that optimizes the stimulation location and orientation based on the gathered motor responses. This closed-loop search is fast and user-independent. \nhttps://t.co/A9cKmpKG3z #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiRPCuXkAYqpBx.jpg"}]}, {"id": "1240901822337953792", "user": "Connect2Brain", "date": "2020-03-20T07:23:43+00:00", "text": "5\nBy the end of the ConnectToBrain project in 2025, we expect to have developed\nnew technology capable of correcting dysfunctional brain networks in several\nbrain disorders with better therapeutic efficacy than state-of-the-art\ntechniques.\nhttps://t.co/4FxkVVTcxM #OHBMx", "media": []}, {"id": "1240902270742671360", "user": "Connect2Brain", "date": "2020-03-20T07:25:30+00:00", "text": "6\nThe project is led by Prof. Ilmoniemi , Prof. Ziemann , and Prof. Romani  and is supported by   #ERCSyG (Grant No 810377) from 2019 to 2025.\n\nInterested in collaborating with us? DM or see: \nhttps://t.co/hrxir5jFY7 #OHBMx", "media": []}]}, {"date": "2020-03-20T07:30:27+00:00", "text": "#OHBMx-21 \u2733 #talk\n\nRoberto Guidotti @robbisg*, A Tosoni, C Sestieri\n\n*Aalto University\n\n\u25b6  Choice- and action-predictive signals properties\n\n#Memory #Methods #Motor", "media": [], "ids": ["1240903516039688192"], "thread": [{"id": "1240903790645182465", "user": "robbisg", "date": "2020-03-20T07:31:33+00:00", "text": "1 #DecisionMaking involves the evaluation of #evidence for a particular #choice and the selection of an associated #action. The PPC and the striatum appear suited for transforming mnemonic evidence into appropriate actions. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETiTam1XgAAM-j3.mp4", "content-type": "video/mp4"}]}, {"id": "1240904284100874241", "user": "robbisg", "date": "2020-03-20T07:33:30+00:00", "text": "2 #Encoding: subjects looked at indoor/outdoor images, that were presented 1,3,5 times to modulate #evidence. #Retrieval: the subjects had to respond if the presented image was shown during encoding or not (old/new), using an eye or finger #movement towards a cue (r/l). \ud83d\udc47 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiTlLGWAAAsdNK.jpg"}]}, {"id": "1240904526619607041", "user": "robbisg", "date": "2020-03-20T07:34:28+00:00", "text": "3 We ran across-subjects #searchlight to decode shared patterns of activity related to choice, action, image type and cue side. Regions showing high #accuracy in choice decoding were #LIPS and #RCau, while motor areas were used to decode #actions.", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiUEGWXQAElvfC.jpg"}]}, {"id": "1240904793746477058", "user": "robbisg", "date": "2020-03-20T07:35:32+00:00", "text": "4 We also investigated how the #evidence modulated the #accuracy. Only two regions were affected by the modulation of evidence: RCau and the LIPS. These regions are crucial for #DecisionMaking. Next, we studied the temporal structure of decision patterns.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiUN9iWAAA7RG0.jpg"}]}, {"id": "1240905076409094144", "user": "robbisg", "date": "2020-03-20T07:36:39+00:00", "text": "5 Using #TemporalDecoding, we used each time point to decode tasks across-subjects. Interestingly, we found that #RCau and #LIPS had an accuracy peak during the #decision phase, while #motor areas began to be informative after the GO Signal. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiUhk9XsAAF89M.jpg"}]}, {"id": "1240905392252686341", "user": "robbisg", "date": "2020-03-20T07:37:55+00:00", "text": "6 We used #TemporalGeneralization to shed light on #RCau role. The #WithinSubject temporal trend suggested an overlap of choice+motor trends, but action-predictive accuracy and generalization matrix suggested that RCau may transform #choice into appropriate #MotorAction #OHBMx \ud83d\udc47", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiU1XnXsAE80wT.jpg"}]}, {"id": "1240969695320313861", "user": "robbisg", "date": "2020-03-20T11:53:26+00:00", "text": "Ops! The figure in 2nd tweet is wrong! It was from a previous similar study... Really sorry, the right one is this \ud83d\udc47! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjPNMgWoAAP-r2.jpg"}]}]}, {"date": "2020-03-20T07:45:20+00:00", "text": "#OHBMx-22 \u2733 #talk\n\nMarijn van Vliet @wmvanvliet*\n\n*Aalto University\n\n\u25b6 Design guidelines for analysis scripts \n\n#Tools", "media": [], "ids": ["1240907259812524032"], "thread": [{"id": "1240907591791886337", "user": "wmvanvliet", "date": "2020-03-20T07:46:39+00:00", "text": "1. We joke about how terrible academic code often is. But as science becomes more dependant on code, it starts to scare me. Here are 7 tips for reducing the chances of having to retract your paper due to bugs in your analysis scripts. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiW0JyWsAAxAos.png"}]}, {"id": "1240907907593633792", "user": "wmvanvliet", "date": "2020-03-20T07:47:54+00:00", "text": "2. Divide and conquer to reduce complexity. Tip 1: cut up that 1000+ lines monstrosity of a script. Tip 2: have a strict division of responsibility between individual steps of your pipeline. Tip 3: make it easy to run a single step on a single subject, as well as run all. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiW-JYXkAAB2Iz.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETiW_fqXkAAgAvO.jpg"}]}, {"id": "1240908381323493376", "user": "wmvanvliet", "date": "2020-03-20T07:49:47+00:00", "text": "3. Never trust someone else's code. Never *ever* trust your own code. Tip 4: Save the intermediate result of each script. Tip 5: Plot everything! Just because the end result looks reasonable does not mean you didn't screw up somewhere in the middle. https://t.co/cecPRy2L5t #OHBMx", "media": []}, {"id": "1240908770743640065", "user": "wmvanvliet", "date": "2020-03-20T07:51:20+00:00", "text": "4. Tip 6: Please, please, please, stop copy/pasting everything. Have a single config file for all parameters. Here is something revolutionary: define your filenames *once*! Have templates for them! This will change your life. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiX5amXsAEMsYA.png"}]}, {"id": "1240909335770935296", "user": "wmvanvliet", "date": "2020-03-20T07:53:35+00:00", "text": "5. Tip 7: stop hoarding files!! Apply https://t.co/p1UsGwvhep to your code folder. Put everything in version-control (git) and then *delete things*. I've seen people dig for half an hour to find the one script that constituted the \"working\" version.", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiYbXkWkAANkG9.jpg"}]}, {"id": "1240909873283547136", "user": "wmvanvliet", "date": "2020-03-20T07:55:43+00:00", "text": "6. Let your code be a testament to your scientific rigor and technological creativity. Read more about the why and how of these 7 tips at https://t.co/2Bti052dlm.\n\nDo you have any tips to share? Reply to this thread and let's discuss! #OHBMx", "media": []}, {"id": "1240912835825356800", "user": "wmvanvliet", "date": "2020-03-20T08:07:29+00:00", "text": "Important note to this tip: please do put everything into version control and back it up. Never really irrevocably delete anything human-written! Stuff it in the endless depths of your git history, push to somewhere safe.", "media": []}, {"id": "1240936014652342272", "user": "wmvanvliet", "date": "2020-03-20T09:39:36+00:00", "text": "Just realized there is a bug in my code example above. Because of course there is :) All code is terrible.", "media": []}]}, {"date": "2020-03-20T08:06:53+00:00", "text": "It\u2019s been an honour to bring you #OHBMx from Aus. @LeonieBorne\n @KoussisNikitas\n @AskDrJeg\n @DrBreaky\n @MegaEJ_Campbell\n signing off. \nThoughts are with our EU colleagues & their families in this challenging time. \nPls take a break before the EU Hub takes the helm.", "media": [], "ids": ["1240912684163334145"], "thread": []}, {"date": "2020-03-20T08:12:32+00:00", "text": "Big thanks to our Australian hub. European hub now taking over with @k7hoven and @baranaydogan chairing. The next presentation should start in a few minutes. #OHBMx", "media": [], "ids": ["1240914106003136513"], "thread": []}, {"date": "2020-03-20T08:16:24+00:00", "text": "#OHBMx-23 \u2733 #talk\n\nGerardo Salvato @gerardosalvato*, L Zapparoli @ZapparoliLaura, M Gandola, E Sacilotto, N Ludwig, M Gargano, T Fazia, G Saetta, P Brugger, E Paulesu, G Bottini\n\n*University of Pavia\n \n\u25b6 Attention to body parts alters thermoregulation in Body Integrity Dysphoria (BID).\n\n#Attention #Disorders #Sensory", "media": [], "ids": ["1240915080549732353", "1240915160530915329"], "thread": [{"id": "1240915302252269570", "user": "gerardosalvato", "date": "2020-03-20T08:17:17+00:00", "text": "(1) Individuals with Body Integrity Dysphoria (BID) typically report an intense desire to have one of their healthy limbs to be amputated. BID has been characterized in terms of absent ownership feeling with preserved ownership judgments #OHBMx", "media": []}, {"id": "1240915652472377344", "user": "gerardosalvato", "date": "2020-03-20T08:18:41+00:00", "text": "(2) It is increasingly recognized that the sense of body part ownership and body part temperature are strictly interconnected. Is the alteration of the limb ownership feeling in BID coupled with an alteration of the limb temperature? #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETieNSZWsAAZVI7.png"}]}, {"id": "1240916066458574849", "user": "gerardosalvato", "date": "2020-03-20T08:20:20+00:00", "text": "(3) We tested 7 individuals with BID seeking for amputation of one leg, and 7 healthy matched-controls. They directed their attention to their 4 limbs, one at a time for 1min in randomized order. We measured the subjects\u2019 limbs temperature using a thermal infrared camera #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiekaYWkAE_66J.jpg"}]}, {"id": "1240916551240482817", "user": "gerardosalvato", "date": "2020-03-20T08:22:15+00:00", "text": "(4) We run LMMs with Side (Left, Right), Limb (Arm, Leg), and Group (BID, Controls) as fixed factors. Random slope: Limb and random intercept: Subject. The delta difference \u2018Attention > Rest\u2019 for each limb was used as dependent variable #OHBMx", "media": []}, {"id": "1240916994897186817", "user": "gerardosalvato", "date": "2020-03-20T08:24:01+00:00", "text": "(5) Results showed that when BID individuals with a desire for amputation involving the leg, focused their attention towards both their lower limbs, a decrease of temperature occurred (interaction Group X Limb (p<.002)) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETifZkkXYAE28Xq.png"}]}, {"id": "1240917756557635585", "user": "gerardosalvato", "date": "2020-03-20T08:27:02+00:00", "text": "(6) Our finding is in line with previous studies showing that unilateral body part disowership is coupled with bilateral lower body part temperature. Is it possible to decrease the desire for amputation in individuals with BID by warming up the affected limb? #OHBMx", "media": []}]}, {"date": "2020-03-20T08:30:45+00:00", "text": "#OHBMx-24 \u2733 #talk\n\nFlorian Beissner @f_beissner*, J Manuel, N F\u00e4rber\n\n*Hannover Medical School\n\n\u25b6 fMRI study of sympathoinhibitory acupuncture effects in the hypothalamus\n\n#Connectivity #Instrumentation #Sensory", "media": [], "ids": ["1240918690746507264"], "thread": [{"id": "1240919214757683201", "user": "f_beissner", "date": "2020-03-20T08:32:50+00:00", "text": "(1) #Acupuncture uses needles to stimulate the nervous system. Effects on nausea and vomiting, hypertension, and inflammation are likely mediated by the autonomic nervous system (ANS), whose control centers lie in the brainstem and hypothalamus \u2192 tough regions for #fMRI. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETihbXvWoAACfFQ.jpg"}]}, {"id": "1240919638055292928", "user": "f_beissner", "date": "2020-03-20T08:34:31+00:00", "text": "(2) Does acupuncture influence ANS centres in a point-specific way? We scanned subjects during continuous electroacupuncture on acupoints ST36 (with) and GB37 (w/o autonomic potential). Intensities were matched. ANS was activated using handgrip, slow breathing and LBNP. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETihycoWoAMEubu.jpg"}]}, {"id": "1240920072144617472", "user": "f_beissner", "date": "2020-03-20T08:36:15+00:00", "text": "(3) Data were analyzed using masked ICA to identify functionally discrete hypothalamic regions followed by dual regression to hypothalamus and brainstem. Check out our #openaccess toolbox for doing this (https://t.co/HSD4VzBw1B, Paper: https://t.co/BozA7JvOhb). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiiLsIWsAA-Sbc.jpg"}]}, {"id": "1240920433077170176", "user": "f_beissner", "date": "2020-03-20T08:37:41+00:00", "text": "(4) Analysis of blood pressure variability (0.1 Hz Mayer waves) showed that #acupuncture at ST36 but not GB37 reduced sympathetic excitability. The effect was driven by the handgrip task, which is known to increase sympathetic activity. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiiZDgWkAAbcCz.jpg"}]}, {"id": "1240920771154903040", "user": "f_beissner", "date": "2020-03-20T08:39:01+00:00", "text": "(5) We also found point-specific effects in the hypothalamus, where functional connectivity between known ANS control centres and different brainstem regions changed differentially for both acupoints. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiixtDXQAEsvjf.jpg"}]}, {"id": "1240922315833163777", "user": "f_beissner", "date": "2020-03-20T08:45:09+00:00", "text": "(6) Comparing our results to animal studies by Longhurst, Tjen-A-Looi, and Li , we found a stunning level of agreement. Thanks to Jorge Manuel and Natalia F\u00e4rber and the guys from  for building the LBNP. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETikOH2XgAAKPAW.jpg"}]}]}, {"date": "2020-03-20T08:47:13+00:00", "text": "#OHBMx-25 \u2733 #talk\n\nFlorian Ph.S Fischmeister @FPFischmeister*, C Cecchetto @Cinzia_Cecchet, V Sch\u00f6pf @NeuroVero\n\n*University of Graz\n\n\u25b6 Separating the Neural Correlates of Olfaction in Health and Disease\n\n#Sensory", "media": [], "ids": ["1240922835469639681"], "thread": [{"id": "1240923007520030720", "user": "FPFischmeister", "date": "2020-03-20T08:47:54+00:00", "text": "1 Human odor perception relies on the tight interplay between the olfactory and the trigeminal systems; most odorants, e.g. peppermint, stimulate both systems simultaneously. It is modulated by breathing patterns, like the sniff, and by expectancy and cognition. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETik3ZaXgAAgx-F.jpg"}]}, {"id": "1240923579463733248", "user": "FPFischmeister", "date": "2020-03-20T08:50:11+00:00", "text": "2 Pseudo-free breathing paradigm (breathing matched to visual) allows to present stimuli (rose \u2013 pure olfactory, CO2 \u2013 pure trigeminal, mint \u2013 both systems) during inhale. Top-down processing controlled by red cross announcing stimulus.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETilCi1XYAEVSII.jpg"}]}, {"id": "1240924106129883137", "user": "FPFischmeister", "date": "2020-03-20T08:52:16+00:00", "text": "3 Bottom-up processes: robust and distinct activation in piriform cortices (rose) but also within the insula and somatosensory cortex (CO2) and dorsolateral prefrontal cortices (mint) \u2013 all only when comparing with breathing pattern; not versus Null. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETilwo-X0AEk9Wo.jpg"}]}, {"id": "1240924472661680128", "user": "FPFischmeister", "date": "2020-03-20T08:53:44+00:00", "text": "4 Anosmics patients show activity comparable to healthy controls for trigeminal stimuli; can use this info to identify or localize the odor (Frasnelli et al. 2007). Only spurious activation within piriform cortex and OFC at very lean threshold for rose odor. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETimD-SWoAEGEh0.jpg"}]}, {"id": "1240925048636088320", "user": "FPFischmeister", "date": "2020-03-20T08:56:01+00:00", "text": "5 Top-down revealed elevated activity within piriform cortex, the insula, dorsolateral prefrontal cortices, and the parahippocampal and entorhinal cortices when odor was announced, but not in inverted contrast. Activation only present when controlled for breathing. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETimdtpXkAUogx6.jpg"}]}, {"id": "1240925540321763330", "user": "FPFischmeister", "date": "2020-03-20T08:57:58+00:00", "text": "6 These results further previous literature and emphasize the importance of the tight control of respiration patterns and cognition as an essential part of olfactory perception.  #OHBMx\n\nThanks to @OHBMequinoX for organizing and stay healthy!\n\nFunding:  (KLI 639)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETim91XXsAYevll.jpg"}]}]}, {"date": "2020-03-20T09:01:39+00:00", "text": "#OHBMx-26 \u2733 #talk\n\nAlexander Bowring @ABNeuroscience*, F Telschow, A Schwartzman, T Nichols @ten_photos\n\n*University of Oxford\n\n\u25b6 Confidence Sets for Cohen\u2019s d Effect Size Images\n\n#Methods", "media": [], "ids": ["1240926466302500864"], "thread": [{"id": "1240926645298569217", "user": "ABNeuroscience", "date": "2020-03-20T09:02:22+00:00", "text": "(1) Statistical testing remains the most popular approach to task-fMRI inference. However, with ample power, even the smallest effects will reach statistical significance. For big fMRI datasets, this leads to almost whole-brain activation, even with stringent correction. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETioKtaXkAELEGS.jpg"}]}, {"id": "1240927066977165312", "user": "ABNeuroscience", "date": "2020-03-20T09:04:02+00:00", "text": "(2) To address this, we developed Confidence Sets (CSs), 3D analogs of confidence intervals. Instead of testing for any activation, the CSs make confidence statements about brain regions where Cohen\u2019s d effect sizes exceed, and fall short of, a given threshold value c. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiobdyXQAIhVg1.jpg"}]}, {"id": "1240927824770412545", "user": "ABNeuroscience", "date": "2020-03-20T09:07:03+00:00", "text": "(3) We tested the method on a range of simulated signals using different sample sizes, across 3000 simulated datasets. In all cases, the coverage (i.e. % of times true thresholded Cohen\u2019s d was contained between upper and lower CSs) stayed close to the 95% nominal target. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETio32QXgAAPvY2.jpg"}]}, {"id": "1240928499667472385", "user": "ABNeuroscience", "date": "2020-03-20T09:09:44+00:00", "text": "(4) We computed CSs for the Human Connectome Project working memory task dataset (N = 80), c = 0.5. The red upper CS in the cerebellum, supramarginal gyrus and areas of the PFC, asserts with 95% confidence a Cohen\u2019s d effect size of at least 0.5 in all these regions.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETipj4zWAAApD8q.jpg"}]}, {"id": "1240929079601307648", "user": "ABNeuroscience", "date": "2020-03-20T09:12:02+00:00", "text": "(5) On the same dataset, we computed a traditional thresholding (voxelwise p < 0.05 FWE, green voxels). These green regions are larger, giving evidence for d > 0, while the red CSs are more specific, giving evidence for a practical effect of d > 0.5. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiqIiNXkAEcz8m.jpg"}]}, {"id": "1240929589523820546", "user": "ABNeuroscience", "date": "2020-03-20T09:14:04+00:00", "text": "(6) CSs combine information about the magnitude and reliability of effect sizes usually given separately in effect estimate and statistic maps. For population fMRI studies, tools to interpret the spatial profile of effects will become increasingly important. Thank you! #OHBMx", "media": []}]}, {"date": "2020-03-20T09:15:19+00:00", "text": "#OHBMx-27 \u2733 #talk\n\nBrieuc Lehmann @BrieucLehmann*, R Henson @rikhens, L Geerligs, S White @SimonWhite83\n\n*University of Oxford\n\n\u25b6 Characterising group-level brain connectivity using exponential random graph models\n\n#Connectivity #Methods #Modeling #Networks", "media": [], "ids": ["1240929904176386048"], "thread": [{"id": "1240930148532342785", "user": "BrieucLehmann", "date": "2020-03-20T09:16:17+00:00", "text": "1\ufe0f\u20e3 In connectivity studies, we typically analyse how the brain's network properties change with a phenotype such as age-group or disease condition. \n\nHow can we describe the properties of a group of networks, while accounting for individual variability?\n\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETirUbdWoAAjnBY.mp4", "content-type": "video/mp4"}]}, {"id": "1240930580369416192", "user": "BrieucLehmann", "date": "2020-03-20T09:18:00+00:00", "text": "2\ufe0f\u20e3 Characterising the group-level structure of brain networks is difficult.\n\nGiven the complex interdependencies inherent in network data, it's not trivial to combine information across networks.\n\n#OHBMx", "media": []}, {"id": "1240931130028756992", "user": "BrieucLehmann", "date": "2020-03-20T09:20:11+00:00", "text": "3\ufe0f\u20e3 The main approaches today are either:\n\na) Construct group-level networks, thus ignoring any individual variability, or \n\nb) Compare graph metrics across individual networks, thus ignoring the features shared between individuals and the codependence between metrics.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETir53sXgAEpQrS.jpg"}]}, {"id": "1240931632837808128", "user": "BrieucLehmann", "date": "2020-03-20T09:22:11+00:00", "text": "4\ufe0f\u20e3 We employ exponential random graph models (ERGMs), a powerful and flexible network model, in a Bayesian multilevel framework to model populations of networks.\n\nWe call this approach multi-BERGM.\n\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETisYx-XkAEpARH.mp4", "content-type": "video/mp4"}]}, {"id": "1240932126599626755", "user": "BrieucLehmann", "date": "2020-03-20T09:24:09+00:00", "text": "5\ufe0f\u20e3 By borrowing information across networks, multi-BERGM yields more precise estimates for the individual network model parameters.\n\nOur framework also provides a description of the overall group network structure via the group-level mean parameter.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETis3t-WkAAtHZ4.png"}]}, {"id": "1240932730021478400", "user": "BrieucLehmann", "date": "2020-03-20T09:26:32+00:00", "text": "6\ufe0f\u20e3 Multi-BERGM can also be used to compare networks from two (or more) groups to nail down which network properties are different between the groups - see our preprint for more details! \n\nhttps://t.co/Sd2nIQEjSa\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETitTX2X0AEbk6e.jpg"}]}]}, {"date": "2020-03-20T09:30:04+00:00", "text": "#OHBMx-28 \u2733 #talk\n\nS\u00e9gol\u00e8ne M. R. Gu\u00e9rin @SegoleneGuerin*, M A Vincent, Y N Delevoye-Turrell @YDelevoye\n\n*University of Lille\n\n\u25b6 fNIRS to Examine Frontal Activity During Whole-body Motor Behaviors: Methodological Issues\n\n#Applications #Methods #Motor", "media": [], "ids": ["1240933615963459584"], "thread": [{"id": "1240933912932646913", "user": "SegoleneGuerin", "date": "2020-03-20T09:31:14+00:00", "text": "1. fNIRS is an imaging method that makes use of light absorption to evaluate cortical haemodynamic responses. This tool can be used in exercise-related protocols given that it is far less sensitive to movement than traditional scanning techniques (e.g., fMRI or EEG). \n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiun_3XkAE7-V7.jpg"}]}, {"id": "1240934582956015621", "user": "SegoleneGuerin", "date": "2020-03-20T09:33:54+00:00", "text": "2. However, scientific community is sorely lacking of validation studies systematically comparing fNIRS suitability in simple vs. whole-body movements in order to confirm the use of this neuroimaging technique to measure brain functions while moving.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETivUjbXgAEk0Jm.jpg"}]}, {"id": "1240935358533156864", "user": "SegoleneGuerin", "date": "2020-03-20T09:36:59+00:00", "text": "3. To address this methodological ellipsis, five participants performed both a circle-drawing task (i.e., arm-only movement) and two cycle-ergometery tasks (i.e., whole-body movement) in synchronisation with a metronome (slow vs. fast pace cuing).\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiv_HFXgAA_r9F.jpg"}]}, {"id": "1240935955093209088", "user": "SegoleneGuerin", "date": "2020-03-20T09:39:21+00:00", "text": "4. Heart and respiratory rates, as well as changes in oxygenated haemoglobin concentrations in both prefrontal and motor areas were recorded. An automatic tracking was used to detect a shift of the fNIRS headset in relation to the participant's head.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiwmPWXgAAUufa.jpg"}]}, {"id": "1240936467985305601", "user": "SegoleneGuerin", "date": "2020-03-20T09:41:24+00:00", "text": "5. The power spectral density analyses indicated that distinct cardiac and respiratory frequency bands were found in the fNIRS data for the two type of motor tasks, which confirm the validity of the recorded fNIRS signals. No headset shift was detected in either task.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETixE6aXYAE1cQM.jpg"}]}, {"id": "1240936978205597696", "user": "SegoleneGuerin", "date": "2020-03-20T09:43:25+00:00", "text": "6. Results encourage the use of physiological recordings to remove accurately frequency bands of physiological noise from the data of interest. Our method provides the means to collect legitimate data during physical activity.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETixZe6XsAAv46a.png"}]}]}, {"date": "2020-03-20T09:47:41+00:00", "text": "#OHBMx-000 \u2733 #break\n\n\u25b6 COFFEE BREAK\n\nTip 1 You can easily find the presentations by searching the IDs (#OHBMx-XX) shown in the program https://t.co/EEsKF4XEGT\n\nTip 2 We make sure that the presenters are available to answer your questions! Go ahead and ask during presentations", "media": [], "ids": ["1240938051934851072"], "thread": []}, {"date": "2020-03-20T10:00:20+00:00", "text": "#OHBMx-29 \u2733 #keynote\n\nLauri Parkkonen @Lauri_Parkkonen*\n\n*Aalto University\n\n\u25b6 KEYNOTE: How much do brain signals tell about brain function? https://t.co/3BFLfBnKkc", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETiy-_mXkAMFgZQ.jpg"}], "ids": ["1240941235474489344"], "thread": [{"id": "1240941377590104070", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:00:54+00:00", "text": "(1) How much is there to measure in the human brain? Very coarse estimate: 100 billion neurons and each could fire 100\u2019s times per second => 10^13\u202614 bits/s (fills up 1-TB hard drive every second!). And this is only the action potentials. #OHBMx", "media": []}, {"id": "1240941775826665472", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:02:29+00:00", "text": "(2) Many methods to measure brain function. Combination of spatial/temporal resolution/field-of-view different in each method but effective data rates surprisingly similar: 10^5\u20266 bits/s (my unpubl. estimates). Thus, we measure less than one millionth of what\u2019s going on! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi1xotWsAAEL5P.jpg"}]}, {"id": "1240942283140325388", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:04:30+00:00", "text": "(3) Our way to measure more: bring #MEG sensors closer to brain by using novel optically-pumped magnetometers (OPM) instead of conventional SQUIDs. https://t.co/H6KlbGdakx #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi2HzdXsAAE2sX.jpg"}]}, {"id": "1240942925028241409", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:07:03+00:00", "text": "(4) Having #MEG sensors right on scalp enables capturing brain activity with higher spatial resolution https://t.co/10eL8h8iYx #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi22gdWkAAhMqR.jpg"}]}, {"id": "1240943579205455872", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:09:39+00:00", "text": "(5) \u2026 and with higher signal-to-noise ratio of e.g. gamma-band activity in early visual cortices https://t.co/d02nhkOKBr #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi3mOGXkAAI9CD.jpg"}]}, {"id": "1240944186431504385", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:12:04+00:00", "text": "(6) This on-scalp vs. conventional #MEG could yield ~3x more information. But that\u2019s still not much wrt what there is to measure (cf 1st tweet) https://t.co/gSctYx9oIJ #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi32k8XYAAq0pr.png"}]}, {"id": "1240944738305540097", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:14:15+00:00", "text": "(7) Is the amount of data really limiting our understanding of brain function? Likely not! Analogue: Try to figure out how a computer works by measuring it without a theory #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi4UqKXgAMpyhw.jpg"}]}, {"id": "1240945264040558592", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:16:21+00:00", "text": "(8) Physiologically-inspired and -constrained models urgently needed to fuse and interpret brain data! Analogue: Successful detection of exoplanets through indirect observations combined with physical models #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi429DWoAAI9vf.png"}]}, {"id": "1240945761052954624", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:18:19+00:00", "text": "(9) Good models in neuroscience: Generative at multiple scales (enabling data fusion), explicitly neurophysiological (interpretable, correctly constrained), parsimonious (simple), and generalizable (useful) #OHBMx", "media": []}, {"id": "1240946195524202496", "user": "Lauri_Parkkonen", "date": "2020-03-20T10:20:03+00:00", "text": "(10) Conclusion: We should continue improving brain measurement methods but even more importantly start constructing computational models that bridge spatiotemporal scales and measurement modalities! Tnx!    #OHBMx https://t.co/40JVUjNL1D", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi5x4BXkAIo8Hv.jpg"}]}]}, {"date": "2020-03-20T10:30:05+00:00", "text": "#OHBMx-30 \u2733 #talk\n \nSofie Valk @sofievalk*, T Xu @TingsterX, D Margulies @DanielMargulies, S Kharabian @shahrzad_kh, C Paquola @CaseyPaquola, A Goulas @AlGoulas, P Kochunov @PKochunov, J Smallwood @the_mindwanders, T Yeo @bttyeo, B Bernhardt @BorisBernhardt,   S Eickhoff \n \n*Forschungszentrum J\u00fclich,  Max Planck Institute for Human Cognitive and Brain Sciences\n \n\u25b6 The Shape of Brain Structure\n \n#Anatomy #Connectivity #Networks", "media": [], "ids": ["1240948723015913473", "1240948824597762050"], "thread": [{"id": "1240948958593081345", "user": "sofievalk", "date": "2020-03-20T10:31:02+00:00", "text": "1. Hello! I will show how we evaluated the genetic basis of natural axes \ud83c\udf08 in the cerebral cortex using cortical thickness covariance (population-based covariance of regional thickness). Our findings might help to understand how the shape of the \ud83e\udde0 enables its function. #OHBMx", "media": []}, {"id": "1240949238625906688", "user": "sofievalk", "date": "2020-03-20T10:32:08+00:00", "text": "2. So, we studied the organization of cortical thickness covariance patterns and its genetic basis through a gradient analysis. See the presentation of  later today to get to know this awesome method and toolbox!\ud83d\udd27> https://t.co/I1KWJdJLTV #OHBMx", "media": []}, {"id": "1240949906027753474", "user": "sofievalk", "date": "2020-03-20T10:34:47+00:00", "text": "3. In the \ud83d\udc6dHuman Connectome Project dataset (& eNKI) we found that structural covariance of thickness is organized along an anterior-to-posterior \u2194\ufe0fand inferior-to-superior \u2195\ufe0faxis. We found the same axes when evaluating the genetic correlation between thickness parcels. #OHMBX", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi9QktXkAAPvmT.jpg"}]}, {"id": "1240950878942760960", "user": "sofievalk", "date": "2020-03-20T10:38:39+00:00", "text": "4. Next, we studied the organization of thickness covariance in macaques \ud83d\udc12(PRIME-DE). Gradient\ud83c\udf08dimensions had similar organizational principles in macaques and humans, suggesting that these axes of \ud83e\udde0organization are phylogenetically conserved within primate species. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi9vBkWkAACxUE.jpg"}]}, {"id": "1240951578284130304", "user": "sofievalk", "date": "2020-03-20T10:41:26+00:00", "text": "5. In \ud83d\udc6d and \ud83d\udc12the inferior-superior \u2195\ufe0fdimension of cortical organization aligned with the dual \ud83d\udc6f-origin theory. Here, cortical areas are conceptualized as waves \ud83c\udf0aof laminar differentiation that spring from the piriform (paleo)-cortex and the hippocampus (archi-cortex). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETi-k1TWsAAf0CJ.jpg"}]}, {"id": "1240952371523575808", "user": "sofievalk", "date": "2020-03-20T10:44:35+00:00", "text": "6. So yeah, to sum up, we show the important role that genetic processes play in determining the large-scale organization of \ud83e\udde0 structure, and so provide an \ud83d\udd0e into the innate architecture supporting human cognition. Preprint\ud83d\udcdc: https://t.co/KEyeNmjHgs #thanks #StayHome #OHBMx \ud83d\udc4b", "media": []}, {"id": "1241034365066457091", "user": "sofievalk", "date": "2020-03-20T16:10:24+00:00", "text": "Oh and... just started my lab  and am recruiting! Drop me a line if you are into this kind of research \ud83c\udf08\ud83e\udde0\ud83d\udc6f\u200d\u2640\ufe0f\ud83d\ude4c", "media": []}]}, {"date": "2020-03-20T10:45:22+00:00", "text": "#OHBMx-31 \u2733 #talk\n \nNicolas Traut @nicolas_traut*, M Fouquet, A Beggiato @anitabeggiato, R Delorme, T Bourgeron @ThomasBourgeron, R Toro @R3RT0\n \n*Pasteur Institute\n \n\u25b6 Grey-White Matter Contrast in Autism: A Replication Study\n \n#Anatomy", "media": [], "ids": ["1240952567057846272"], "thread": [{"id": "1240952903403212800", "user": "nicolas_traut", "date": "2020-03-20T10:46:42+00:00", "text": "1\nThe contrast of the interface between the cortical grey and white matter is emerging as an important neuroimaging phenotype. In autism spectrum disorder (ASD), the study of Andrews et al (2017) showed a significant decrease of this contrast in several areas. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjAE8oXYAARFWl.jpg"}]}, {"id": "1240953439909224448", "user": "nicolas_traut", "date": "2020-03-20T10:48:50+00:00", "text": "2\nWe tried to replicate this study in two large cohorts: ABIDE 1 and 2 (Autism Brain Imaging Data Exchange, N=1,477) and EU-AIMS (European Autism Interventions \u2013 A Multicentre Study for Developing New Medications, N=586). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjAf1EWkAAbk2O.jpg"}]}, {"id": "1240953916658987014", "user": "nicolas_traut", "date": "2020-03-20T10:50:44+00:00", "text": "3\nWe did find a statistically significant result in the ABIDE cohort, although the direction of the effect was the opposite: instead of a reduction we found an increase of grey-white matter contrast in the ASD group. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjA_0EXYAEtXAL.jpg"}]}, {"id": "1240954708103180291", "user": "nicolas_traut", "date": "2020-03-20T10:53:52+00:00", "text": "4\nWe then tried to replicate the effect in the EU-AIMS cohort, but we did not find anything significant. Upon closer examination of the ABIDE analysis, we made a concerning observation: the statistically significant effect was due only to 1 site: NYU (New York University) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjBsHNWkAA_GAY.jpg"}]}, {"id": "1240955041218998272", "user": "nicolas_traut", "date": "2020-03-20T10:55:12+00:00", "text": "5\nAdditionally, we looked at cortical thickness differences. We observed again a statistically significant effect: increased cortical thickness among subjects with ASD. Again, this was exclusively driven by the NYU data. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjB-oLXQAEVpZv.jpg"}]}, {"id": "1240955437593264128", "user": "nicolas_traut", "date": "2020-03-20T10:56:46+00:00", "text": "6\nUntil the cause for this difference is discovered, the NYU data should be treated with special attention. In order to facilitate tracking this artefact down, we have made our code available on GitHub: https://t.co/oozI9Svwbs. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjCMqcWkAAve9D.png"}]}]}, {"date": "2020-03-20T11:00:22+00:00", "text": "#OHBMx-32 \u2733 #talk\n \nKatja Heuer @katjaQheuer*, M Kleineberg @marian42_, R Dinnage @ecologician, C C Sherwood @chet_sherwood, W D Hopkins, E Schwartz, G Langs @georg, R Valabregue, M D Santin, M Herbin, R Toro @R3RT0\n \n*Max Planck Institute for Human Cognitive and Brain Sciences ()\n \n\u25b6 BrainScapes: The landscape of possible primate brain shapes\n \n#Anatomy", "media": [], "ids": ["1240956343168765959", "1240956441466413056"], "thread": [{"id": "1240956479047434246", "user": "katjaQheuer", "date": "2020-03-20T11:00:55+00:00", "text": "1\nThe shape of primate brains varies widely from small smooth to profusely folded large brains. Studying morphological diversity across phylogeny allows us to better understand how primate brains adapt, and in particular the evolutionary context of the human brain.\n#ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjDNb2XsAAbcQX.jpg"}]}, {"id": "1240958200083554304", "user": "katjaQheuer", "date": "2020-03-20T11:07:45+00:00", "text": "2\nRecent advances in generative machine learning models have led to algorithms capable of learning shape embeddings and to generate realistic new instances. We explored an autoencoder deep neural network to generate shapes of primate brains including 34 different species\n#ohbmx", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1240957409411170305/pu/vid/1440x720/PS3Ktyj6n4Wy8u_n.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1240959317844656134", "user": "katjaQheuer", "date": "2020-03-20T11:12:11+00:00", "text": "3\nOur network successfully learnt a landscape of changes in shape. Interestingly, species with brains of comparable volume were close in the learnt space, despite having been size-normalised for the training. \nImage1:t-SNE embedding of latent space learned by autoencoder\n#ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjFmNBWoAE-IG4.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETjFxs9WoAAtvKh.jpg"}]}, {"id": "1240960024341594112", "user": "katjaQheuer", "date": "2020-03-20T11:15:00+00:00", "text": "4\nThis suggests that changes in volume are consistently concomitant with changes in shape, and this disregarding the species\u2019 position in the phylogenetic tree. \nHere, we show the Mantel test for phylogeny\u2013brainVolume and latentSpace\u2013BrainVolume.\n#ohbm", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjGf1dXgAARx-D.png"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETjGimKXgAETaxC.png"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETjGjiQWkAE9r7s.png"}]}, {"id": "1240960668125270017", "user": "katjaQheuer", "date": "2020-03-20T11:17:33+00:00", "text": "5\nOur network is also able to generate new data. We generated possible brain shapes for all ancestral states of the primate phylogenetic tree, and evolutionary trajectories for each of our species back to the brain of the common ancestor. \n#ohbmx", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1240960405188612096/pu/vid/720x720/blvDv9ZnpB8nQZxf.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1240961309207912448", "user": "katjaQheuer", "date": "2020-03-20T11:20:06+00:00", "text": "6\nWe successfully trained a deep neural network to learn the space of morphological variation across primates, and to generate new data along the phylogenetic tree. We obtained evolutionary trajectories of extant primate brains all the way back to the common ancestor.\n#OHBMx", "media": []}, {"id": "1240962327018721291", "user": "katjaQheuer", "date": "2020-03-20T11:24:09+00:00", "text": "Soooooooo many thanks to all the incredible team! This is a true pleasure with all of you \u2764\ufe0f\ud83d\udc12\u2764\ufe0f\ud83d\udc12\ud83d\udc35\ud83e\udd70 @R3RT0 @marian42_ @chet_sherwood @ecologician   @georg Bill Hopkins Romain Valabregue Ernst Schwartz Marc Herbin \ud83d\udc12\ud83d\udc35", "media": []}, {"id": "1240962888220782594", "user": "ecologician", "date": "2020-03-20T11:26:23+00:00", "text": "Thank you Katja! Great job!", "media": []}]}, {"date": "2020-03-20T11:15:28+00:00", "text": "#OHBMx-33 \u2733 #talk\n \nRoberto Toro @R3RT0*, K Heuer @katjaQheuer\n \n*Pasteur Institute\n \n\u25b6 The development of brain folding patterns, real and ideal.\n \n#Anatomy #Developmental You can follow the presentation at https://t.co/ZSdcu0qhr0", "media": [], "ids": ["1240960143694745603", "1240962314859315200"], "thread": [{"id": "1241334611910758402", "user": "R3RT0", "date": "2020-03-21T12:03:29+00:00", "text": "Oops! I forgot to reply...\n\"The developmentt of brain folding patterns, real and ideal\", together with @katjaQheuer, is here:\nhttps://t.co/Vkq2DbhvjX", "media": []}]}, {"date": "2020-03-20T11:27:34+00:00", "text": "Unfortunately, the next presentation has been cancelled. So it's an extra coffee break! Feel free to grab a coffee, ask more questions from our brilliant talks so far or just browse #OHBMx. See you again at 11.45 UTC!", "media": [], "ids": ["1240963189073985536"], "thread": []}, {"date": "2020-03-20T11:45:13+00:00", "text": "#OHBMx-35 \u2733 #talk\n \nPascal Vrticka @PVrticka*, T Nguyen @trinh_nguyen9, H Schleihauf @HannaSchleihauf, E Kayhan, D Matthes, S Hoehl @HoehlStefanie\n \n*University of Essex\n \n\u25b6 The effects of interaction quality on neural synchrony during mother-child problem solving\n \n#Developmental #Social", "media": [], "ids": ["1240967628287606786", "1240967678749298694"], "thread": [{"id": "1240967890951643137", "user": "PVrticka", "date": "2020-03-20T11:46:15+00:00", "text": "(1) Mutual attunement of behavior and physiology between #children & #caregivers plays a vital role for #attachment and the development of #social and #emotional competences. Such bio-behavioral #synchrony can be observed on (at least) four different levels outlined below. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjNsOwXQAAy_Cb.jpg"}]}, {"id": "1240968244401451008", "user": "PVrticka", "date": "2020-03-20T11:47:40+00:00", "text": "(2) In this study, we focused on two of the above levels: #behavior and #brain #activity. In N=42 #mother-#child dyads (child age 5 years), we video-recorded behavior and measured brain activity using functional near-infrared spectroscopy (#fNIRS) #hyperscanning. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjN0R5WoAEtJ2y.jpg"}]}, {"id": "1240968767154290688", "user": "PVrticka", "date": "2020-03-20T11:49:44+00:00", "text": "(3) Mother-child dyads performed a problem-solving task (#tangram #puzzle). They either did so together (1 puzzle for both; #cooperation condition) or separated by an opaque screen (2 puzzles, 1 for each; #individual condition). There was an additional #rest condition. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjOKXDXkAAFPNK.jpg"}]}, {"id": "1240969321175814146", "user": "PVrticka", "date": "2020-03-20T11:51:56+00:00", "text": "(4) We observed highest #neural #synchrony during the #cooperation condition, and neural synchrony positively correlated with behavioral task performance (number of templates solved). This finding suggests that neural coupling is important to #social #information exchange. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjOrQuX0AA9EX2.jpg"}]}, {"id": "1240969774471024641", "user": "PVrticka", "date": "2020-03-20T11:53:44+00:00", "text": "(5) During #cooperation, #neural #synchrony also positively correlated with behavioral #reciprocity (contingent responses resulting in a turn-taking quality as behavioral flow) and #child #agency (child was able to engage in the task instead of being led by the #mother). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjPH8UWAAA08m8.jpg"}]}, {"id": "1240970299501424644", "user": "PVrticka", "date": "2020-03-20T11:55:50+00:00", "text": "(6) Our results emphasize #neural #synchrony as a #biomarker for #mother-#child #interaction #quality. Ongoing studies are examining #father-child dyads and use several self-report and narrative-based assessments of #attachment in both parents and children. Thank you \ud83d\ude00\ud83e\udde0 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjPgigWAAAsKpJ.jpg"}]}]}, {"date": "2020-03-20T12:00:01+00:00", "text": "#OHBMx-000 \u2733 #break\n \n\u25b6 COFFEE BREAK\n \nThank you to our presenters for a fascinating session. Time for a coffee. Feel free to engage with the presenters or browse #OHBMx\n \nSee you in 15!", "media": [], "ids": ["1240971355279052800"], "thread": []}, {"date": "2020-03-20T12:15:07+00:00", "text": "#OHBMx-36 \u2733 #talk\n\nSarah Faber @sciencebanshee*, A McIntosh @ar0mcintosh\n\n*University of Toronto\n\n\u25b6 Modelling Musical Improvisation\n\n#Modeling #Networks #Social", "media": [], "ids": ["1240975155377160192"], "thread": [{"id": "1240975446596030466", "user": "sciencebanshee", "date": "2020-03-20T12:16:17+00:00", "text": "1. WHO WANTS TO TALK MUSIC?! In this review paper, we mapped out the top-down creative and bottom-up sensory/perceptual aspects of music-making, creating a conceptual model (screen reader-compatible version here: https://t.co/in5Pfba5Uy). Let\u2019s dive in! #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjUlrVWoAAnEkU.mp4", "content-type": "video/mp4"}]}, {"id": "1240975895000764417", "user": "sciencebanshee", "date": "2020-03-20T12:18:04+00:00", "text": "2. The model! The first 2 tiers are Auditory Processing and Output Monitoring. Sound hits the ears (music has sound, don\u2019t @ me) and is processed by regions in the bilateral auditory cortex before being sent to higher-order regions for syntactic and semantic processing. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjUu1WWoAEwf8I.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETjUw27XkAU1LWK.jpg"}]}, {"id": "1240976249830457346", "user": "sciencebanshee", "date": "2020-03-20T12:19:28+00:00", "text": "3. We wanted to EXPRESSLY detail these perceptual networks in the model because if you\u2019re improvising, you\u2019re listening to/monitoring your output to make sure you sound the way you want to sound. Also, this sound-> meaning+error is VERY similar to how language works. #OHBMx", "media": []}, {"id": "1240976666429657088", "user": "sciencebanshee", "date": "2020-03-20T12:21:08+00:00", "text": "4. The next tier is for top-down production and planning. Here, you make a plan, then you physically DO that plan. There\u2019s an extra bit here (Flow State) involving the dorsolateral and orbito-frontal prefrontal cortex that deactivate with highly skilled improvisers. WHY. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjVhnjX0AY7NJA.jpg"}]}, {"id": "1240976929802588162", "user": "sciencebanshee", "date": "2020-03-20T12:22:10+00:00", "text": "5. The final tier is for social improvisation. Interestingly, the dorsolateral prefrontal cortex is active here in expert improvisers, indicating extra cognitive control is needed when another person is in the mix. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjV0X2WoAAgeBz.jpg"}]}, {"id": "1240977515272900609", "user": "sciencebanshee", "date": "2020-03-20T12:24:30+00:00", "text": "6. There's so much more to discuss, so check out our paper: https://t.co/cPbLGRDeyD\n\n(Pre-print versions: https://t.co/No7yRHDyYt)\n\nMany thanks to @OHBMequinoX for their organization and for YOU for turning up! Let\u2019s chat in the comments! #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjWeEwXYAAnSGY.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T12:30:05+00:00", "text": "#OHBMx-37 \u2733 #talk\n\nXenia Kobeleva @dr_xenia*\n\n*University Hospital Bonn\n\n\u25b6 Clinical neuroimaging- where art thou? Tutorial on clinically relevant research questions \n\n#Neurology #Psychiatry #Tools", "media": [], "ids": ["1240978921782489095"], "thread": [{"id": "1240979190498963457", "user": "dr_xenia", "date": "2020-03-20T12:31:09+00:00", "text": "1. Complex neuroimaging tools &methods might impede communication between researchers and clinicians. This might lead to less focussed research and more research \"waste\". This is a tutorial on how to create questions in neuroimaging that are relevant for clinical research. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjX5FGXkAE69aQ.mp4", "content-type": "video/mp4"}]}, {"id": "1240979926624305154", "user": "dr_xenia", "date": "2020-03-20T12:34:05+00:00", "text": "2. The PICO system from evidence-based medicine helps to translate clinical problems into well-formulated questions (Sackett et al. 2000; doi:10.7326/ACPJC-1995-123-3-A12; https://t.co/k0BQZQe6QS). It divides the questions into four parts: #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjYhjvWoAUfGBX.jpg"}]}, {"id": "1240980608454713345", "user": "dr_xenia", "date": "2020-03-20T12:36:48+00:00", "text": "3. While formulating the whole questions with all four parts, you should focus on one component in your research, i.e. on the population. Neuroimaging might help with selecting the right population for a given intervention or diagnostic procedure. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjZFY9WoAAJG9Q.jpg"}]}, {"id": "1240981243350614021", "user": "dr_xenia", "date": "2020-03-20T12:39:19+00:00", "text": "4. Alternatively you could use neuroimaging to improve diagnostic procedures or interventions (by targeting regions). It is important to compare them to the current standard (i.e. hippocampus volume vs. new complex functional connectivity algorithm). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjZX1EXsAA0x0a.jpg"}]}, {"id": "1240981794192834560", "user": "dr_xenia", "date": "2020-03-20T12:41:30+00:00", "text": "5. Neuroimaging is also a great tool to evaluate outcomes (e.g., in clinical trials). Also in this case, it should be compared to standard outcome predictions using clinical examinations. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjZ-UrWoAkeTs2.jpg"}]}, {"id": "1240982193243119618", "user": "dr_xenia", "date": "2020-03-20T12:43:05+00:00", "text": "6. All in all, make sure your question is relevant for clinical research (talk to clinicians!) and more than just one more publication. And while formulating the whole PICO question, focus on improving one of the PICO components with your neuroimaging method! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjaki1XYAc6ynd.jpg"}]}]}, {"date": "2020-03-20T12:45:09+00:00", "text": "#OHBMx-38 \u2733 #talk\n\nAmelie Haugg @amhaugg*, M Habegger, A Speckert, S Meier, R Sladky @sweetneuron, P Staempfli, C Lor, E van Maren, A Watve, A Manoliu, E Seifritz, M Kirschner @MMG_Kirschner, M Herdener, B B Quednow, F Scharnowski @FrankScharnowsk\n\n*University of Zurich\n \n\u25b6 Adaptive neurofeedback stimulation to support smoking cessation\n\n#Disorders #Psychiatry", "media": [], "ids": ["1240982711914946560", "1240982772245843968"], "thread": [{"id": "1240983268796903424", "user": "amhaugg", "date": "2020-03-20T12:47:22+00:00", "text": "(1) Controlling one\u2019s cigarette craving is a key factor to quit smoking. In this study, we used a novel paradigm combining neuro-adaptive cue exposure and fMRI neurofeedback (NFB) to help smokers tolerate cigarette craving better and to support their smoking cessation. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjbsAyWoAMU7Hz.mp4", "content-type": "video/mp4"}]}, {"id": "1240983763812782080", "user": "amhaugg", "date": "2020-03-20T12:49:20+00:00", "text": "(2) 64 smokers who wanted to quit or reduce their cigarette consumption participated in the study. They trained either their craving-driven anterior cingulate cortex (ACC; experimental group, EG, N=32) or their non craving-related angular gyrus (control group, CG, N=32). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjb8cTWkAEGQs5.jpg"}]}, {"id": "1240984241682472960", "user": "amhaugg", "date": "2020-03-20T12:51:14+00:00", "text": "(3) Subjects in the EG had to downregulate their ACC activity. They were presented stimuli whose craving intensity was dynamically coupled to ongoing ACC activity. The better subjects were at downregulating, the more intense got the presented smoking cues, and vice versa. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjcZWiXsAAPtJr.mp4", "content-type": "video/mp4"}]}, {"id": "1240984997760294912", "user": "amhaugg", "date": "2020-03-20T12:54:14+00:00", "text": "(4) Subjects in the CG received the same instructions, but intensity of the presented cues was linked to the angular gyrus. Both groups were trained for 10 NFB runs within 2 sessions, and came back for a follow up after 6 weeks. #OHBMx", "media": []}, {"id": "1240985499143155712", "user": "amhaugg", "date": "2020-03-20T12:56:14+00:00", "text": "(5) At the follow-up session, subjects in the EG showed more reduction in cigarette consumption (p<0.05) and Fagerstr\u00f6m dependence scores (p<0.01) as compared to before training than subjects in the CG. The EG reduced 40 cigarettes/week, the CG 15 cigarettes/week. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjda3cWAAAwHJ7.jpg"}]}, {"id": "1240986013020966916", "user": "amhaugg", "date": "2020-03-20T12:58:16+00:00", "text": "(6) Our results suggest that adaptive NFB stimulation can support smoking cessation by reducing craving when being confronted with nicotine cues. Consequently, brain-controlled adaptive cue exposure stimulation might be promising novel therapeutic tool in addiction. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjd07PXgAUzwHS.jpg"}]}, {"id": "1240986346233225217", "user": "amhaugg", "date": "2020-03-20T12:59:36+00:00", "text": "Thanks to all the amazing people who were involved in this study!\n\nAnd thank you @OHBMequinoX for organizing such a fantastic conference full of neuroscience distraction!", "media": []}]}, {"date": "2020-03-20T13:00:06+00:00", "text": "#OHBMx-39 \u2733 #talk\n\nFabian Renz @Fabian_M_Renz*, D Steyrl, A Haugg, C Lor, S J G\u00f6tzendorfer, A A Nicholson, R Sladky, S Skouras, A McDonald, C Craddock, L Hellrung, M Kirschner, M Herdener, Y Koush, J Keynan, T Hendler, K Cohen Kadosh, C Zich, M Papoutsi, J MacInnes,   A Adcock, K Dickerson, N-K Chen, K Young, J Bodurka, M Marxen, Y Shuxia, B Becker, T Auer, R Schweizer, G Pamplona, R A Lanius, K Emmert, S Haller, D Van De Ville, D-Y Kim, J-H Lee, T Marins, M Fukuda, B Sorger, T Kamp, S-L Liew, R Veit, M Spetter, N Weiskopf,   F Scharnowski\n\n*University of Vienna\n\n\u25b6 Predicting neurofeedback training performance\n\n#Methods #Modeling #Tools", "media": [], "ids": ["1240986475631726592", "1240986535220125696", "1240986625733201922"], "thread": [{"id": "1240987695779852288", "user": "Fabian_M_Renz", "date": "2020-03-20T13:04:57+00:00", "text": "1.\nReal-time fMRI-based neurofeedback is an emerging scientific and clinical tool that allows for learning to self-regulate brain activity. It has the capacity to modulate the behaviour of healthy individuals and was successfully applied to various clinical populations.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjfQ8cXQAYC9NZ.jpg"}]}, {"id": "1240988922722889728", "user": "Fabian_M_Renz", "date": "2020-03-20T13:09:50+00:00", "text": "2.\nHowever, self-regulation performance varies considerably across runs and individuals as illustrated in Figure 2. Here, we investigate whether neurofeedback performance across runs are merely random or follow predictable patterns. \n #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjghhwWsAAYewF.png"}]}, {"id": "1240989633334382592", "user": "Fabian_M_Renz", "date": "2020-03-20T13:12:39+00:00", "text": "3.\nWe employed a meta-analytic approach including data from 11 studies. Using machine learning, we tried to predict regulation performance (measured by % signal change) based on previous run performances and study/participant specific information. \n #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjhLIkWsAgBw_O.png"}]}, {"id": "1240990149934284801", "user": "Fabian_M_Renz", "date": "2020-03-20T13:14:42+00:00", "text": "4.\nPredictions were always better than chance level. The best performance was achieved for predicting the 4. run (median R\u00b2 = 0.26). However, the inclusion of study/participant information did not improve explained variance.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjh26PWoAgRSdw.png"}]}, {"id": "1240990561739386887", "user": "Fabian_M_Renz", "date": "2020-03-20T13:16:21+00:00", "text": "5.\nFurthermore, the contributions of individual features were analysed using permutation-based feature importance. It turned out that the most important predictors were previous performances, with the most recent run being the most important predictor in 3 out of 4 cases.\n#OHBMx", "media": []}, {"id": "1240990934168412160", "user": "Fabian_M_Renz", "date": "2020-03-20T13:17:49+00:00", "text": "6.\nBased on this meta analytic approach investigating the predictability of real-time fMRI-based neurofeedback it can be concluded that neurofeedback training is not random but reveals predictable patterns. \n#OHBMx", "media": []}]}, {"date": "2020-03-20T13:15:16+00:00", "text": "#OHBMx-40 \u2733 #talk\n\nLeonie Henschel @deepmilab*, S Conjeti, S Estrada, K Diers, B Fischl, M Reuter\n\n*German Center for Neurodegenerative Diseases (DZNE)\n\n\u25b6 FastSurfer - A fast and accurate deep learning based neuroimaging pipeline\n\n#Applications #Methods #Modeling #Tools", "media": [], "ids": ["1240990289428385792"], "thread": [{"id": "1240990443501953024", "user": "deepmilab", "date": "2020-03-20T13:15:52+00:00", "text": "(1) FastSurfer https://t.co/y8xNPswZdv is a fast and extensively validated neuroimaging pipeline based on . We combine an advanced neural network (FastSurferCNN) with a surface pipeline for the automated processing of structural human brain MRIs.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjiMp8WAAAN_7e.jpg"}]}, {"id": "1240990991915593729", "user": "deepmilab", "date": "2020-03-20T13:18:03+00:00", "text": "(2) FastSurferCNN segments 95 structures in <1 min (GPU). It is composed of three F-CNNs operating on coronal, axial, sagittal 2D slice stacks (7-channel) and a final view aggregation combining the advantages of 3D patches (local neighbourhood) and 2D slices (global view).\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjii3SX0AA7cnt.jpg"}]}, {"id": "1240991479172128768", "user": "deepmilab", "date": "2020-03-20T13:19:59+00:00", "text": "(3) FastSurferCNN outperforms other DL networks by a significant margin both w.r.t. FreeSurfer & a manual standard (best DSC & Hausdorff distance for subcortical & cortical structures). It generalizes well to downsampled & defaced images, across vendors & disease states.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETji1JdXkAE9u2k.jpg"}]}, {"id": "1240991996317229056", "user": "deepmilab", "date": "2020-03-20T13:22:03+00:00", "text": "(4) FastSurfer is highly reliable as demonstrated by the close agreement between the thickness and volumetric measurements for 20 Test-Retest subjects from OASIS1 (average ICC of 0.9 on cortical and 0.99 on subcortical regions).\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjjP1RXkAEHwKc.jpg"}]}, {"id": "1240992572102901768", "user": "deepmilab", "date": "2020-03-20T13:24:20+00:00", "text": "(5) FastSurfer shows high reproducibility of known disease effects: it robustly detects reduced cortical thickness in regions associated with dementia as well as subcortical volume differences with increased sensitivity relative to FreeSurfer.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjjt2vWkAMPUNj.jpg"}]}, {"id": "1240993493683765249", "user": "deepmilab", "date": "2020-03-20T13:28:00+00:00", "text": "(6) FastSurfer provides @FreeSurferMRI conform outputs, enables time-critical clinical applications and scalable big-data analysis. Follow us for code release notice https://t.co/W8s9kRiYE5 and visit our webpage https://t.co/sJyjUv9WtO for job offers. Thank you :)!\n#OHBMx", "media": []}]}, {"date": "2020-03-20T13:30:05+00:00", "text": "#OHBMx-41 \u2733 #talk\n\nAnees Abrol @aneesabrol*, S Plis @PlisSergey, V Calhoun @vdcalhoun\n\n*TReNDS Center Atlanta\n\n\u25b6 How to beat deep learning with any method: the art of apples and oranges!\n\n#Applications #Connectivity #Methods #Tools", "media": [], "ids": ["1240994020706398208"], "thread": [{"id": "1240994404841684994", "user": "aneesabrol", "date": "2020-03-20T13:31:37+00:00", "text": "1/6 DL methods may be hyped, may not necessarily perform better and may not scale boundlessly with more data on every task. Yet, it is important to benchmark by using what we already know about their applicability (e.g. in representation learning). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjlpjgXQAMoSEY.jpg"}]}, {"id": "1240994768844455938", "user": "aneesabrol", "date": "2020-03-20T13:33:04+00:00", "text": "2/6 We demonstrate on an age+gender classification task that DL provides more predictive features and shows asymptotic behavior like standard ML (SML), but with significantly higher performance. SML can perform equally well if trained on representations learnt by DL models #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjmA0dXQAANaT2.jpg"}]}, {"id": "1240995391006314497", "user": "aneesabrol", "date": "2020-03-20T13:35:32+00:00", "text": "3/6 DL presents lower empirical asymptotic complexity in relative computational time: While DL methods are notorious for high computational time complexity, high empirical computational costs for CPU-based SML implementations for high n are often overlooked. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjmW0bWoAAJfRm.jpg"}]}, {"id": "1240996000250179584", "user": "aneesabrol", "date": "2020-03-20T13:37:57+00:00", "text": "4/6 DL Embeddings Span a Comprehensible Projection Spectrum: Representational patterns of the brain are indeed learnt; these distil continually with increasing training data and eventually evolve into separate gender clusters, both presenting a gradual spectrum of age. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjm4jzX0AARQa2.jpg"}]}, {"id": "1240996563448709126", "user": "aneesabrol", "date": "2020-03-20T13:40:11+00:00", "text": "5/6 DL Enables Robust Prediction Relevance Estimates for Brain Regions: High similarity was observed in saliency maps (a) across independently sampled repetitions and (b) saliency methods (gradient-based backpropagation and network occlusion sensitivity analysis). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjnhvJXYAMBDZh.png"}]}, {"id": "1240997223254700032", "user": "aneesabrol", "date": "2020-03-20T13:42:49+00:00", "text": "6/6 Future work must focus on utilizing the predictive power of DL encodings & facilitating methodical model interpretation. Rather than focusing on ways to show DL does not predict well in some cases, their flexibility should be leveraged to solve problems that SML cannot #OHBMx", "media": []}]}, {"date": "2020-03-20T13:44:05+00:00", "text": "#OHBMx-000 \u2733 #break\n \n\u25b6 COFFEE BREAK\n\nAnother great session at the #OHBMx conference. That concludes almost 12 hours of awesome presentations and #neuroscience discussions! Let us thank our presenters once again! Next talk starts at 14.00 UTC, till then https://t.co/xOmxT4pZkD", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjor0TWAAAeAdJ.mp4", "content-type": "video/mp4"}], "ids": ["1240997544701853697"], "thread": []}, {"date": "2020-03-20T14:00:00+00:00", "text": "#OHBMx-42 \u2733 #keynote\n\nKarla Miller @fmrib_karla*\n\n*Wellcome Centre for Integrative Neuroimaging, University of Oxford\n\n\u25b6 KEYNOTE: MRI meets microscopy: tools and techniques for multi-scale neuroscience https://t.co/cMAlmBVjjy", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjm3lmXgAAQiWR.jpg"}], "ids": ["1241001547972202496"], "thread": [{"id": "1241002072960573442", "user": "fmrib_karla", "date": "2020-03-20T14:02:05+00:00", "text": "1 Brains span scales: function requires structures spanning many orders of magnitude. Individual neurons communicate by sending molecules across tiny gaps; regions communicate using long-range connections 10 million X larger. Neuroscience needs multi-scale tools! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjso9aXkAE7UYY.jpg"}]}, {"id": "1241002564927315968", "user": "fmrib_karla", "date": "2020-03-20T14:04:02+00:00", "text": "2 No single tool can span these scales. So our group combines MRI (mm) with light microscopy (um) and electron microscopy (nm). Post-mortem MRI is a crucial intermediary: same contrast as in-vivo MRI and same tissue/state as microscopy. But it\u2019s not as easy as it sounds\u2026 #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjs9bpWoAMvhr0.jpg"}]}, {"id": "1241003349241139200", "user": "fmrib_karla", "date": "2020-03-20T14:07:09+00:00", "text": "3 Post-mortem whole brain is \ud83d\ude31 for MRI: T2 & diffusion plummet. We developed acquisitions that cope. Here: diffusion SSFP@7T: 0.5mm in 5 days (!). Less ambitiously, 0.85mm in a few hrs. Work of Sean Foxley &  https://t.co/cKMxDL91bk https://t.co/6nD23W8o1N #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjt5oNWkAYf-8q.jpg"}]}, {"id": "1241003889186521089", "user": "fmrib_karla", "date": "2020-03-20T14:09:18+00:00", "text": "4 We need to adapt signal models too. Here: we translate  SSFP to spin echo equivalent. This model accounts for transmit inhomogeneity @7T and non-Gaussianity to give a well-defined b-value. Work with  &  https://t.co/jUOCiDBvIL https://t.co/CEZKoXew6R #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjuOpjWkAM1uiD.jpg"}]}, {"id": "1241005017710067713", "user": "fmrib_karla", "date": "2020-03-20T14:13:47+00:00", "text": "5 Another big problem: registration. Tissue distorts & tears during microscopy prep. And how do we find where it came from in the brain? Enter the reg wizards! TIRL is a new tool soon to be released in FSL by  & Mark Jenkinson https://t.co/3mqSxzK9Jx  #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjvee3XQAc__yb.jpg"}]}, {"id": "1241005197738065921", "user": "fmrib_karla", "date": "2020-03-20T14:14:30+00:00", "text": "6 We are starting to map multi-modal MRI to multi-modal histopathology in ALS. Whole-brain sampling uses known progression patterns to sample a range of pathological severity. If we can learn relationships, we can predict histopathology from MRI! https://t.co/Fa5Q3wCyXs #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjvmDUXgAATAIU.jpg"}]}, {"id": "1241005777210486786", "user": "fmrib_karla", "date": "2020-03-20T14:16:48+00:00", "text": "7 Microscopy can inform our MRI tools. Jeroen Mollink validated WM fibre dispersion estimates from diffusion agree well with microscopy. PLI+MRI also has potential to inform sophisticated diffusion tractography, e.g. to overcome the gyral bias (R). https://t.co/hy3WSHByxU #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjv7T-XQAMm0nK.jpg"}]}, {"id": "1241006393311846403", "user": "fmrib_karla", "date": "2020-03-20T14:19:15+00:00", "text": "8 We can jointly model MRI and microscopy data from the same tissue. Here, we use microscopy to overcome a degeneracy in MRI: dispersion & radial diffusion look the same. We showed that FRFs vary across the brain. Work by  & @SaadJbabdi https://t.co/CYCFO3ByBP #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjwYXAWkAAuk3J.jpg"}]}, {"id": "1241007000412213249", "user": "fmrib_karla", "date": "2020-03-20T14:21:40+00:00", "text": "9 At finer scale, we can model subcellular structures from 3D EM data. Fresh out TODAY by Michiel Kleinnijenhuis: semi-automated algorithms turn 3DEM of WM into 3D meshes of 6 compartments for realistic diffusion simulations. https://t.co/xvSmB79WqL #ohbmx https://t.co/W1siKZsZrc", "media": []}, {"id": "1241007391178661889", "user": "fmrib_karla", "date": "2020-03-20T14:23:13+00:00", "text": "10 But we don\u2019t have All The Ideas - you can play too! Behold BigMac: one brain with in-vivo dMRI & rs-fMRI, ex-vivo dMRI & whole-brain microscopy. 1000 directions@3 shells! PLI & more! Curated by @AmyFDHoward. Open resource coming soon. Sneak peek: https://t.co/IIYn7B8DGm #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjxeY7WkAMIujf.jpg"}]}, {"id": "1241007691163750409", "user": "fmrib_karla", "date": "2020-03-20T14:24:25+00:00", "text": "Thanks @OHBMequinoX for what could well be the best neuroimaging conference of 2020! To express my gratitude, here\u2019s Nemo trying to hide in a bed of axons because reasons. (Credit Michiel Kleinnijenhuis) #ohbmx https://t.co/eORkjT5abm", "media": []}]}, {"date": "2020-03-20T14:30:24+00:00", "text": "#OHBMx-43 \u2733 #talk\n\nAngela Tam @angelatamtweets*, J Chen @JianzhongChen9, V Kebets @valeria_kebets, L Q R Ooi @Leon_Oo1, S Marek @smarek0502, N Dosenbach @ndosenbach, S Eickhoff @INM7_ISN, D Bzdok @danilobzdok, A Holmes @AvramHolmes, B T Yeo @bttyeo\n \n*National University of Singapore\n\n\u25b6 Shared & unique network features predict cognition, mental health & personality in kids\n\n#Connectivity #Developmental #Modeling", "media": [], "ids": ["1241009197963259904", "1241009332894019585"], "thread": [{"id": "1241009551698063361", "user": "angelatamtweets", "date": "2020-03-20T14:31:48+00:00", "text": "1) We can use functional connectivity (FC) from rest & task states to predict individual differences in behavior. Most studies predicted cognition & some found that task outperforms rest. Can we use FC to predict a wider range of behaviors & find their neural substrates? #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETjzmrcUMAAEL9D.mp4", "content-type": "video/mp4"}]}, {"id": "1241009759769071621", "user": "angelatamtweets", "date": "2020-03-20T14:32:38+00:00", "text": "2) In the ABCD dataset, we used kernel ridge regression with rest-FC & task-FC, from 3 tasks (Monetary Incentive Delay (MID), Stop Signal Task (SST), N-back), from 400 cortical and 19 subcortical ROIs to predict 36 behaviors spanning cognition, personality & mental health. #OHBMx", "media": []}, {"id": "1241009906305478661", "user": "angelatamtweets", "date": "2020-03-20T14:33:13+00:00", "text": "3) We tested 5 models: 4 single-kernel (rest, MID, SST, N-Back) & 1 multi-kernel (Multi-kernel FC, a combination of all 4 brain states). On average, task-FC outperforms rest for cognition but not other behaviors & combining rest & task improves cognition & personality. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETjz52yU8AEEc6b.jpg"}]}, {"id": "1241010107921530880", "user": "angelatamtweets", "date": "2020-03-20T14:34:01+00:00", "text": "4) The prediction of related behaviors was supported by similar brain networks that were distinct from other behavioral domains. E.g. cognition was predicted by similar features that were different from personality & mental health. This was consistent across brain states. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj0C6bUYAAyxK_.jpg"}]}, {"id": "1241010494250430470", "user": "angelatamtweets", "date": "2020-03-20T14:35:33+00:00", "text": "5) So which networks support the prediction of cognition, personality & mental health? We did a conjunction analysis of the significantly predictive connections that were common across the 4 brain states for the 3 behavioral domains. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj0b7ZUMAAr7QL.jpg"}]}, {"id": "1241010966130601984", "user": "angelatamtweets", "date": "2020-03-20T14:37:25+00:00", "text": "6) Cognition was notably predicted by somatomotor & salience, personality by default & dorsal attention, & mental health by default & control networks. In short, related behaviors are broadly supported by shared patterns of brain organization. Thanks for listening! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj0xN9VAAAf7Tb.jpg"}]}]}, {"date": "2020-03-20T14:45:12+00:00", "text": "#OHBMx-44 \u2733 #talk\n\nBradly Alicea @balicea1*, S Dvoretskii, Z Gong, J Parent @JesParent, A Gupta\n\n*Orthogonal Research and Education Lab\n\n\u25b6 Developmental Braitenberg Vehicles\n\n#Connectivity #Developmental #Modeling #Tools", "media": [], "ids": ["1241012923755102212"], "thread": [{"id": "1241013106496716802", "user": "balicea1", "date": "2020-03-20T14:45:56+00:00", "text": "1  #Neuroethology is the study of brains and behavior in the context of an animal\u2019s natural environment. This can be specified and controlled in simulated environments through interdependencies among brain, body, and environment #OHBMx", "media": []}, {"id": "1241013699478945792", "user": "balicea1", "date": "2020-03-20T14:48:17+00:00", "text": "2  We use developmental Braitenberg Vehicles (dBVs) to study neuronal and behavioral development. Different types of environment are utilized to simulate and learn about the varieties of dBV developmental experience (Images: Thomas Schoch, Wikimedia) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj3W58X0AEmV3y.png"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETj3W6GX0AQqN-7.png"}]}, {"id": "1241014368046854151", "user": "balicea1", "date": "2020-03-20T14:50:56+00:00", "text": "3  One approach to dBVs involves using genetic algorithms to produce a pattern of nervous system connectivity. The fitness function is based on triangulating prior movement vectors, and results in sensorimotor-driven spatial learning #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj35-yX0AEJ-j6.jpg"}]}, {"id": "1241015033708138497", "user": "balicea1", "date": "2020-03-20T14:53:35+00:00", "text": "4  A second approach is to use a multisensory (olfactory, gustatory) model of the environment. Individual dBVs explore such environments, while their brains use Hebbian learning to encode information and develop nervous system connectivity #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj4l0dX0AAw0dY.png"}]}, {"id": "1241015784241074176", "user": "balicea1", "date": "2020-03-20T14:56:34+00:00", "text": "5  The third approach to dBVs involve BV collectives. A homogeneous population of vehicles respond to a common stimulus, which produces large-scale response patterns. These patterns can be understood as extended neural phenotypes #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj5QtSXsAAJqps.jpg"}]}, {"id": "1241016396013862912", "user": "balicea1", "date": "2020-03-20T14:59:00+00:00", "text": "6  Interested in learning more? Then check out the preprint from  live now on ! There you will learn about our open source software packages, and more information about dBVs: https://t.co/NVkLb0T6PK #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj50ojWkAAsqLT.jpg"}]}, {"id": "1241016471029010433", "user": "balicea1", "date": "2020-03-20T14:59:18+00:00", "text": "7  Thank you! #OHBMx", "media": []}]}, {"date": "2020-03-20T15:00:01+00:00", "text": "#OHBMx-45 \u2733 #talk\n\nRuby Kong @rubykong92*, Y R Tan @YanRui36453910, S Harrison, J Bijsterbosch @PersonomicsLab, B Bernhard @BorisBernhardt, S Eickhoff @INM7_ISN, B T Yeo @bttyeo\n\n*National University of Singapore\n \n\u25b6 Comparing gradients and parcellations for RSFC behavioral prediction\n\n#Connectivity #Modeling", "media": [], "ids": ["1241016651237203969", "1241016701787004929"], "thread": [{"id": "1241016973711953920", "user": "rubykong92", "date": "2020-03-20T15:01:18+00:00", "text": "1. Gradients versus parcellations for functional connectivity prediction of behavioral. Who wins? Here, we compare 2 gradient techniques, 2 soft-parcellation techniques and 2 hard-parcellation techniques. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj6UBYUUAEt80v.jpg"}]}, {"id": "1241017182512803841", "user": "rubykong92", "date": "2020-03-20T15:02:07+00:00", "text": "2. We used rs-fMRI data from 746 HCP subjects to obtain individual-specific  features representing functional connectivity for different gradient and parcellation approaches. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj6f_hUYAAQHA9.jpg"}]}, {"id": "1241017338582798338", "user": "rubykong92", "date": "2020-03-20T15:02:45+00:00", "text": "3. Kernel ridge regression (KRR) was used to predict each behavior. The input to KRR is an inter-subject similarity matrix. KRR relies on the idea that the behavior of a test subject is more similar to the behavior of a training subject if their brains are more similar. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj6rhQVAAEJBJP.jpg"}]}, {"id": "1241017514789728257", "user": "rubykong92", "date": "2020-03-20T15:03:27+00:00", "text": "4. For each approach, the inter-subject similarity matrix was constructed based on individual-specific features representing functional connectivity. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj60I9U0AEhkuB.jpg"}]}, {"id": "1241017688836599810", "user": "rubykong92", "date": "2020-03-20T15:04:08+00:00", "text": "5. We performed 20-fold cross-validation: KRR was trained on 19 folds and used to predict behavior in the test fold. Regularization parameters were determined using inner-loop cross-validation. The 20-fold cross-validation was repeated 100 times to ensure stability. #OHBMx", "media": []}, {"id": "1241017992458039298", "user": "rubykong92", "date": "2020-03-20T15:05:21+00:00", "text": "6. In a KRR framework, RSFC from brain parcellations outperformed gradients for behavioral prediction. Albeit gradient techniques are relatively new, so future research might yield additional improvements. Further testing in new datasets is also desirable. Thank you! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj7I5jUcAEX1D_.png"}]}]}, {"date": "2020-03-20T15:14:55+00:00", "text": "#OHBMx-46 \u2733 #talk\n\nVanessa Teckentrup @glassybrain*, S Neubert @neuro_bird, T Kircher, A Krug, I Nenadi\u0107, D Grotegerd, U Dannlowski, M Walter, N B Kroemer @cornu_copiae\n\n*University of T\u00fcbingen\n \n\u25b6 Insular connectivity profiles in depression\n\n#Connectivity #Disorders #Networks #Psychiatry", "media": [], "ids": ["1241020403096653824", "1241020475737886731"], "thread": [{"id": "1241020567819558912", "user": "glassybrain", "date": "2020-03-20T15:15:35+00:00", "text": "(1) Sensing their bodily state is often difficult for people with depression. Such imbalances in interoceptive vs exteroceptive processing have been associated with signaling in the insular cortex. But what makes the insula special? #OHBMx https://t.co/dusj6Izpdx", "media": []}, {"id": "1241020958435086341", "user": "glassybrain", "date": "2020-03-20T15:17:08+00:00", "text": "(2) Distinct subregions for interoceptive and exteroceptive processing have been shown in the insula. So, if we measure the insula with fMRI at rest, can we find evidence for an impaired segregation between insular subregions in depression? #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj93YKWAAIjIKK.jpg"}]}, {"id": "1241021436535410698", "user": "glassybrain", "date": "2020-03-20T15:19:02+00:00", "text": "(3) We calculated similarity (correlation) and density (shift function) for 6 insular subregions (anterior vs posterior) based on resting-state fMRI in N=850. A greater similarity in functional connectivity (FC) profiles would then indicate a loss of segregation. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj-JAgXgAAII-H.jpg"}]}, {"id": "1241021948454469632", "user": "glassybrain", "date": "2020-03-20T15:21:04+00:00", "text": "(4) Patients suffering from depression showed less distinct FC profiles compared to healthy participants, indicative of less segregation. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj-hssWkAEd_9q.jpg"}]}, {"id": "1241022460205715457", "user": "glassybrain", "date": "2020-03-20T15:23:06+00:00", "text": "(5) Crucially, we found a greater similarity in FC profiles across insular subregions in depression which was specific to the insula and driven by the anterior insula. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETj-_hRXsAEiI8X.jpg"}]}, {"id": "1241022956752515075", "user": "glassybrain", "date": "2020-03-20T15:25:04+00:00", "text": "(6) In sum, we found an impaired segregation of functionally distinct signaling in the insula. This increased reliance on interoceptive signals linked to the anterior insula might contribute to the pathophysiology in depression. #OHBMx https://t.co/fqvgIpYzyr", "media": []}]}, {"date": "2020-03-20T15:30:10+00:00", "text": "#OHBMx-47 \u2733 #talk\n\nSara Larivi\u00e8re @saratheriver*, R Rodr\u00edguez-Cruces @rcruces, M E Caligiuri, A Gambardella, L Concha, S Keller @BrainImagingUoL, F Cendes, C Yasuda, R K\u00e4lvi\u00e4inen, G D Jackson, M Kowalczyk, M Semmelroch, M Severino, P Striano, D Tortora, S Hatton,   E Epilepsy Working Group , C D Whelan , P Thompson , S M Sisodiya, A Bernasconi, A Labate, C McDonald, N Bernasconi, B C Bernhardt \n\n*Montreal Neurological Institute\n \n\u25b6 Network-based atrophy modelling in the common epilepsies\n\n#Connectivity #Disorders #Modeling #Networks", "media": [], "ids": ["1241024239760822276", "1241024363002040320", "1241024470535503872"], "thread": [{"id": "1241024857002905602", "user": "saratheriver", "date": "2020-03-20T15:32:37+00:00", "text": "1. Epilepsy is increasingly conceptualized as a network disorder. Here we will show how we identified network mechanisms underlying atrophy patterns in 1,021 epileptic patients relative to 1,564 healthy controls from 19 international sites. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETkBXH8WAAA8Lxy.mp4", "content-type": "video/mp4"}]}, {"id": "1241025350571786240", "user": "saratheriver", "date": "2020-03-20T15:34:35+00:00", "text": "2. Using mixed effects models, we first established patterns of atrophy in temporal lobe epilepsy (TLE) and idiopathic/genetic generalized epilepsy (GE). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkB6qBWkAEKXvI.jpg"}]}, {"id": "1241025687353528320", "user": "saratheriver", "date": "2020-03-20T15:35:55+00:00", "text": "3. Next, we evaluated whether these abnormalities were guided by normative network organization. We found that while cortical hubs were most affected in TLE, cortico-subcortical hubs were most vulnerable in GE. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkCQJnXsAEflW8.jpg"}]}, {"id": "1241026027897397248", "user": "saratheriver", "date": "2020-03-20T15:37:16+00:00", "text": "4. We then localized specific disease epicenters in mesiotemporal and limbic cortices in TLE as well as sensorimotor cortices and subcortical areas in GE. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkCi_nWkAApwH2.jpg"}]}, {"id": "1241026451664707584", "user": "saratheriver", "date": "2020-03-20T15:38:57+00:00", "text": "5. Lastly, we assessed markers of disease progression and showed a strong influence of connectome architecture on how the disease unfolds in TLE. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkC7nMXQAI37D9.jpg"}]}, {"id": "1241026754078298118", "user": "saratheriver", "date": "2020-03-20T15:40:09+00:00", "text": "6. Through worldwide collaboration in @enigmabrains-Epilepsy, we offer a novel network-based mechanistic perspective that can help to understand the pathological cascades in the common epilepsies. Thank you! #OHBMx", "media": []}, {"id": "1241053227606593536", "user": "BrainImagingUoL", "date": "2020-03-20T17:25:21+00:00", "text": "\ud83d\udc4f \ud83d\udc4f", "media": []}]}, {"date": "2020-03-20T15:45:09+00:00", "text": "#OHBMx-48 \u2733 #talk\n\nSarah Morgan @Sarah_Morgan_UK*, J Seidlitz @jakob_seidlitz, K Whitaker @kirstie_j, R Romero-Garcia @rafa_romero_gar, N Clifton @NE_Clifton, C Scarpazza @CScarpazza, T van Amelsvoort, M Marcelis, J van Os @JimvanOs1, G Donohoe @GaryD2011,   D Mothersill , A Corvin, A Pocklington, A Raznahan , P McGuire, P Vertes , E Bullmore \n\n*Cambridge University\n \n\u25b6 Cortical patterning of abnormal morphometric similarity in psychosis\n\n#Connectivity #Disorders #Networks #Psychiatry", "media": [], "ids": ["1241028008502333445", "1241028049103093760", "1241028139914010625"], "thread": [{"id": "1241028440217858051", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:46:51+00:00", "text": "We investigated structural brain connectivity in psychosis, using a new method called \u2018morphometric similarity mapping\u2019. We used 3 independent MRI case-control datasets, collated by @PSYSCAN. Cases had psychotic disorders (mainly schizophrenia), controls were healthy 1/6 #OHBMx", "media": []}, {"id": "1241028952342265856", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:48:54+00:00", "text": "Morphometric similarity was proposed by @jakob_seidlitz & quantifies similarity between brain regions in terms of their structure (thickness, curvature etc). @jakob_seidlitz showed structurally similar regions are more likely connected by tracts https://t.co/y1Uqr71Svk 2/6 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkFJsBXkAA-qrC.jpg"}]}, {"id": "1241029262787989504", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:50:08+00:00", "text": "We found that morphometric similarity was globally reduced in psychosis patients in all 3 of our datasets, implying that patients\u2019 brain regions were more differentiated from each other and therefore potentially less interconnected 3/6 #OHBMx", "media": []}, {"id": "1241029551058235392", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:51:16+00:00", "text": "Regionally, morphometric similarity was especially decreased in frontal and temporal regions. Interestingly, these are the regions with the highest morphometric similarity in healthy subjects, i.e. \"hub\" regions. Again, results broadly replicated across the 3 studies 4/6 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkFshyWAAEz6e4.jpg"}]}, {"id": "1241030092802973699", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:53:26+00:00", "text": "We then combined our MRI map of brain abnormalities with brain-wide gene expression data from . We found a small cluster of genes highly expressed in the brain regions most vulnerable to psychosis & up-regulated in prior post-mortem schizophrenia studies 5/6 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkF-hfWsAANYU8.jpg"}]}, {"id": "1241030506831065090", "user": "Sarah_Morgan_UK", "date": "2020-03-20T15:55:04+00:00", "text": "Overall, we begin to see how combining genomics and imaging can give a more integrative understanding of schizophrenia. More details at: https://t.co/l4e6aZoNtS, including links to code and data. Thanks to all co-authors and study participants, and to you for reading! 6/6 #OHBMx", "media": []}]}, {"date": "2020-03-20T16:00:51+00:00", "text": "#OHBMx-000 \u2733 #break\n\n\u25b6 COFFEE BREAK\n\nHave a break, have a chit chat. A virtual one.\n\nYou can discuss anything related under #OHBMx. Or check out the program and read abstracts at https://t.co/CGpVqTCbqd", "media": [], "ids": ["1241031961147322368"], "thread": []}, {"date": "2020-03-20T16:15:07+00:00", "text": "#OHBMx-49 \u2733 #talk\n\nYaren Y\u0131lmaz @yarenyilmazz*, M \u015eanl\u0131er @MugeSanlier, P D B Top\u00e7ular\n\n*European Academy of Neurology, Turkish Neurological Society\n\n\u25b6 Amyloid Imaging Findings in Familial and Sporadic Patients with Alzheimer\u2019s Disease\n\n#Disorders #Methods #Neurology", "media": [], "ids": ["1241035552809418759"], "thread": [{"id": "1241035838617792512", "user": "yarenyilmazz", "date": "2020-03-20T16:16:15+00:00", "text": "1)Alzheimer\u2019s disease(AD) is a multifactorial dementia disorder characterized by early amyloid-\u03b2, tau deposition, glial activation and neurodegeneration. There are differences between Familial and Sporadic AD.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkLgOWXkAEZZsn.png"}]}, {"id": "1241036342844325888", "user": "yarenyilmazz", "date": "2020-03-20T16:18:16+00:00", "text": "2)Method: Clinical and imaging data of 17 familial and 18 sporadic Alzheimer's patients, that have no difference between the two groups in terms of sociodemographic characteristics and disease duration, from AD Neuroimaging Initiative(ADNI) were included in the study.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkLrU9XgAA7A5O.png"}]}, {"id": "1241036767240826881", "user": "yarenyilmazz", "date": "2020-03-20T16:19:57+00:00", "text": "3)We analysed amyloid PET imaging with VINCI (\u201cVolume Imaging in Neurological Research, Co-Registration and ROIs included\u201d). Flutemetamol radionuclide(F-18) marking amyloid PET images were used.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkMGImWAAEzR4x.png"}]}, {"id": "1241037295844757505", "user": "yarenyilmazz", "date": "2020-03-20T16:22:03+00:00", "text": "4)There was no significant difference between the two groups in terms of total amyloid burden. In the familial group, the amyloid burden was higher in the insular cortex, striatum, supra marginal gyrus, orbitofrontal cortex and cingulate cortex than in the sporadic group.#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkM0AeWAAE4bco.jpg"}]}, {"id": "1241038203035045888", "user": "yarenyilmazz", "date": "2020-03-20T16:25:39+00:00", "text": "5)The findings of our study support other studies suggesting that frontal and extrapyramidal amyloid burden is higher in familial AD cases compared to sporadic cases.#OHBMx \nThank you for reading! \ud83e\udde0", "media": []}]}, {"date": "2020-03-20T16:30:08+00:00", "text": "#OHBMx-50 \u2733 #talk\n\nHenrietta Howells @ettah*, L Vigano @lucavigan2, G Puglisi @guglielmo_psi, A Leonetti @leonetti_23, M Rabuffetti, L Simone, A Bellacicca @ABellacicca, L Bello, L Fornia @ForniaLuca, G Cerri\n\n*University of Milan\n \n\u25b6 Mapping tracts for complex hand movement in awake neurosurgery\n\n#Anatomy #Disorders #Motor #Neurology", "media": [], "ids": ["1241039332217180161", "1241039450894893056"], "thread": [{"id": "1241039674816303104", "user": "ettah", "date": "2020-03-20T16:31:30+00:00", "text": "1\ufe0f\u20e3Imagine your body stops moving as it should. After #neurosurgery, higher order motor deficits can be as disabling as hemiplegia. Surgeons balance between preserving function & #tumour removal but need guidance to avoid resecting important regions and connections #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETkO7wdX0AA9nOx.mp4", "content-type": "video/mp4"}]}, {"id": "1241040611014213635", "user": "ettah", "date": "2020-03-20T16:35:13+00:00", "text": "2\ufe0f\u20e3Electrical stimulation \u26a1\ufe0f induces but also stops #hand movement \ud83d\udc4c in awake patients in certain frontal regions. This is useful for indicating where to preserve to reduce motor deficits. It has let us examine functional properties of #cortex but also the white matter-s! #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkPrzlX0AIlG0D.jpg"}]}, {"id": "1241041447765377031", "user": "ettah", "date": "2020-03-20T16:38:33+00:00", "text": "3\ufe0f\u20e3To see if stimulating different frontal #whitematter tracts changes effects on ongoing hand movement ->> 36 patients with intraoperative stim during \ud83d\udc4c manipulation + #BCBToolkit #disconnectome analysis + preop #tractography #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkQMKYWAAAfmNA.jpg"}]}, {"id": "1241042010032594947", "user": "ettah", "date": "2020-03-20T16:40:47+00:00", "text": "4\ufe0f\u20e3We quantified the effect of #whitematter stimulation measuring change in phasic muscle activity with EMG. Subcortical \u26a1\ufe0f either stopped the ongoing action or changed muscle coordination required to perform it, with medial->lateral anatomical distinctions #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETkRG-BXgAE6dvW.mp4", "content-type": "video/mp4"}]}, {"id": "1241042664532967425", "user": "ettah", "date": "2020-03-20T16:43:23+00:00", "text": "5\ufe0f\u20e3Different combinations of tracts were associated with different stim effects using #disconnectome. Group level #tractography showed arrest linked to connections of superior frontal gyrus while loss of coordination most associated with lateral parieto-frontal tracts #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkRtVEXkAM6_r-.jpg"}]}, {"id": "1241042998319812610", "user": "ettah", "date": "2020-03-20T16:44:42+00:00", "text": "6\ufe0f\u20e3Our results may reflect disruption of different hierarchical circuits for #hand motor control. Thank you to our surgical & research team \ud83d\udc68\u200d\u2695\ufe0f\ud83d\udc69\u200d\ud83c\udfeb\ud83d\udc68\u200d\ud83c\udfebhttps://t.co/ZXuBti4PZ2 & to the organisers. And thank you for reading! Preprint imminent. Any Qs at all, please tweet/DM me #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETkSAkXXYAEwVQh.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T16:45:13+00:00", "text": "#OHBMx-51 \u2733 #talk\n\nDominik Krzemi\u0144ski @dokatox*, N Masuda, K Hamandi, K D Singh @magnetokrishio, B Routley @BethRoutley, J Zhang @ccbrainlab\n\n*Cardiff University Brain Research Imaging Centre \n\u25b6 Pairwise Maximum Entropy Model for MEG resting state data - the epilepsy study\n\n#Disorders #Methods #Networks", "media": [], "ids": ["1241043127345057793", "1241043259675357184"], "thread": [{"id": "1241043556963409920", "user": "dokatox", "date": "2020-03-20T16:46:56+00:00", "text": "(1/5) Hello everyone \ud83d\ude42 In this study we applied the Energy landscape analysis (based on the Ising model from statistical physics) to define instantaneous state dynamics of common resting-state networks: DMN, FPN, SMN. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkSaTtXkAQI6a5.jpg"}]}, {"id": "1241043707673141248", "user": "dokatox", "date": "2020-03-20T16:47:32+00:00", "text": "(2/5) In #neuroscience, the method is more known as pairwise maximum entropy model, and has been applied previously to spikes, LFP, or fMRI data. Our study is the first - to our knowledge - to extend it to M/EEG. #OHBMx", "media": []}, {"id": "1241044016587776003", "user": "dokatox", "date": "2020-03-20T16:48:45+00:00", "text": "(3/5) To estimate energy landscape of the fast-varying signal one needs to (a) calculate 'Hilbert transform' envelopes in frequency bands (b) binarise the signal (c) fit parameters of the pMEM model (d) identify local minima #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkS1OpWsAgFT8B.png"}]}, {"id": "1241044511259795458", "user": "dokatox", "date": "2020-03-20T16:50:43+00:00", "text": "(4/5) In our work, the energy landscape is calculated from 6-min. #meg resting-state recordings of healthy subjects and patients suffering from juvenile myoclonic #epilepsy (JME). The model has good fitting accuracy, due to the high temporal resolution of the data. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkTDZwWAAA5rMv.png"}]}, {"id": "1241044769234661376", "user": "dokatox", "date": "2020-03-20T16:51:45+00:00", "text": "(5/5) JME group had fewer local energy minima than controls and also elevated energy values for the FPN in theta, beta and gamma. The proportion of time the FPN was occupied within the basins of energy minima was shortened among JME patients. More: https://t.co/kHt7GTYZlP #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkTnNuWAAEaS1c.png"}]}]}, {"date": "2020-03-20T17:00:04+00:00", "text": "#OHBMx-52 \u2733 #talk\n\nThomas Hinault @T_Hinault*, S Courtney\n\n*Inserm U1077\n\n\u25b6 Aging and dynamic network connectivity\n\n#Connectivity #Networks", "media": [], "ids": ["1241046863903670273"], "thread": [{"id": "1241046974260019204", "user": "T_Hinault", "date": "2020-03-20T17:00:30+00:00", "text": "1/ Hi everyone! My work is about how EEG network synchrony and DTI structural integrity to understand the different cognitive trajectories during healthy #aging: Why are some individuals as cognitively efficient as younger adults while others show some cognitive decline? #OBMHx", "media": []}, {"id": "1241047426145943553", "user": "T_Hinault", "date": "2020-03-20T17:02:18+00:00", "text": "2/ High-density #EEG was recorded in young and older participants during a task targeting working memory and inhibition. Phase-locking value (#PLV) of source EEG was assessed. #DTI were collected in another session. #Graph theory analyses were conducted on both modalities. #OBMHx", "media": []}, {"id": "1241047900349751296", "user": "T_Hinault", "date": "2020-03-20T17:04:11+00:00", "text": "3/ Behavioral results showed a smaller benefit of working memory updating for subsequent interference resolution in older than young adults (compared to maintenance). However, variations were observed in older adults (although similar in age, gender, SES, GM atrophy). #OBMHx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkWSrHWkAEJvfz.jpg"}]}, {"id": "1241048455889502208", "user": "T_Hinault", "date": "2020-03-20T17:06:24+00:00", "text": "4/Results revealed delayed modulations of alpha and gamma #PLV between the right IFG and occipital lobe in older than young adults during cue updating. PLV was mediating the association between individual #DTI integrity (IFO) and behavioral interference levels! #OBMHx #aging", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkWle3XgAAtki6.jpg"}]}, {"id": "1241048984950640640", "user": "T_Hinault", "date": "2020-03-20T17:08:30+00:00", "text": "5/ Altered alpha graph clustering and variability across time was observed in individuals with larger interference. EEG graph clustering was associated with individual DTI graph clustering values, suggesting a strong link between time-varying #EEG and structural #network. #OBMHx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkXGB2XYAA99S2.jpg"}]}, {"id": "1241049631640948736", "user": "T_Hinault", "date": "2020-03-20T17:11:04+00:00", "text": "6/ Results were replicated with data from the CAMCAN database, a preprint is on BioRxiv (https://t.co/Lk2qdJPgW2). Another paper on whole-brain connectivity is in preparation. Please ask if you have questions! And thanks for joining in these troubled times #OBMHx #Aging\ud83d\ude09", "media": []}]}, {"date": "2020-03-20T17:15:13+00:00", "text": "#OHBMx-53 \u2733 #talk\n\nGiulia Bert\u00f2 @GiuliaBert1*, D Bullock @Is_Dan_Bull, P Astolfi, S Hayashi @soichih, L Zigiotto, L Annicchiarico, F Corsini, A De Benedictis, S Sarubbo @SSarubbo, F Pestilli @furranko, P Avesani, E Olivetti @0l1v3tt1\n \n*Bruno Kessler Foundation - Trento - Italy\n\n\u25b6 Classifyber, a supervised algorithm for white matter bundle segmentation\n\n#Anatomy #Methods #Tools", "media": [], "ids": ["1241050675565531136", "1241050770683895808"], "thread": [{"id": "1241050949394849793", "user": "GiuliaBert1", "date": "2020-03-20T17:16:18+00:00", "text": "1. Obtaining accurate segmentation of #whitematter bundles from #tractography #dMRI data in the human brain is essential for multiple applications, such as surgical planning \ud83d\udc68\u200d\u2695\ufe0f in #neurosurgery and group studies \ud83e\uddd0 in cognitive #neuroscience. #OHBMx", "media": []}, {"id": "1241051207520718854", "user": "GiuliaBert1", "date": "2020-03-20T17:17:20+00:00", "text": "2. Although notable improvements have occurred over the years, the quality of automatic segmentation methods is not yet satisfactory, especially when dealing with datasets with very diverse characteristics (e.g. different data quality, tracking algorithms or bundle sizes). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkZdkFXQAADVwU.jpg"}]}, {"id": "1241051456343588864", "user": "GiuliaBert1", "date": "2020-03-20T17:18:19+00:00", "text": "3. To bridge the gap, we  propose a novel supervised streamline-based segmentation method, called Classifyber, which accurately predicts whether or not a given streamline belongs to the bundle of interest, by leveraging example bundles segmented by experts. #ML\ud83d\udca1#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkZqi7XsAc_M_D.jpg"}]}, {"id": "1241051926919348224", "user": "GiuliaBert1", "date": "2020-03-20T17:20:11+00:00", "text": "4. Classifyber combines in a linear model both the similarity measures between streamlines, typical of streamline-based methods, and the anatomical information from Regions of Interest (ROIs), typical of ROI-based and connectivity-based methods. \ud83e\udde0 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkZ7W5WkAE2xox.jpg"}]}, {"id": "1241052413135654915", "user": "GiuliaBert1", "date": "2020-03-20T17:22:07+00:00", "text": "5. Classifyber provides evidence to substantially improve the quality of segmentation with respect to state-of-the-art methods (e.g. TractSeg, RecoBundles, and LAP) and, more importantly, it is robust to different data settings. \ud83e\udd29 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkaUr-XQAgH4SF.jpg"}]}, {"id": "1241052915193741316", "user": "GiuliaBert1", "date": "2020-03-20T17:24:07+00:00", "text": "6. Check our preprint out! https://t.co/wh3egkwY3i\n\nClassifyber is freely available on the online platform  both as the full algorithm implementing training and test phases, and as a pre-trained method.\nhttps://t.co/zcn7VPeib9\nhttps://t.co/xaUlaaaODg\n\nThanks!\ud83d\ude00#OHBMx", "media": []}, {"id": "1241098289824620544", "user": "furranko", "date": "2020-03-20T20:24:25+00:00", "text": "Thanks for sharing this! @OHBMequinoX", "media": []}]}, {"date": "2020-03-20T17:30:25+00:00", "text": "#OHBMx-000 \u2733 #break\n\n\u25b6 THE LONG DARK TEA TIME OF THE SOUL\n\nIt would be lovely if you\u2019d take some time to leave us organisers feedback at https://t.co/7semMizUpP (or even via DMs if you have specific ideas for improvement)! We\u2019ll be back at 17:45 UTC with a keynote by @lauradata", "media": [], "ids": ["1241054500850450434"], "thread": []}, {"date": "2020-03-20T17:44:53+00:00", "text": "#OHBMx-54 \u2733 #keynote\n\nLaura Lewis @lauradata*\n\n*Boston University\n\n\u25b6 KEYNOTE: Fast multimodal neuroimaging of brainwide dynamics across sleep and wakefulness https://t.co/c2cWjZ227w", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkc__SXgAA_xTi.jpg"}], "ids": ["1241058143280549890"], "thread": [{"id": "1241058896049045504", "user": "lauradata", "date": "2020-03-20T17:47:53+00:00", "text": "1. Excited to be part of #OHBMx, thanks to the organizers! Our goal is to understand how the brain transforms its function across sleep and wakefulness. Almost every aspect of cognition, behavior, physiology shifts \u2013 how does this happen? We image neural dynamics to study this", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkgZ75WsAIeWi-.jpg"}]}, {"id": "1241059427521871872", "user": "lauradata", "date": "2020-03-20T17:49:59+00:00", "text": "2. A big challenge is that we want to study time-varying dynamics (100s of milliseconds) across the whole brain. fMRI gives great 3D images of the whole brain, but has classically been considered slow. So \u2013 how fast can fMRI signals really be?  https://t.co/AwZ7oJRRzh  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkg6sDXgAIVDCl.jpg"}]}, {"id": "1241059670963470337", "user": "lauradata", "date": "2020-03-20T17:50:57+00:00", "text": "3. We used oscillating visual stimuli to create known neural oscillations and measured with fast (TR<400 ms) fMRI. We found neural oscillations up to 0.75 Hz, that were ten times larger than predicted by conventional models. So, fMRI signals can be surprisingly fast!   #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkhJOSXsAA9mSb.png"}]}, {"id": "1241060284430778368", "user": "lauradata", "date": "2020-03-20T17:53:24+00:00", "text": "4. Why are fMRI signals so fast? Lots of great work has shown that fMRI signals are nonlinear. We found throughout the visual system that fast neural activity drives faster hemodynamic responses, even faster in deep brain (thalamus, brainstem): https://t.co/xmA9NkW4LJ   #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkhggmWoAIS9iu.jpg"}]}, {"id": "1241060754889093122", "user": "lauradata", "date": "2020-03-20T17:55:16+00:00", "text": "5. With this info, we wanted to study what happens during sleep. Using simultaneous EEG and fast fMRI we imaged neural dynamics during onset of light sleep. Spectral analysis on the fMRI signals shows broad cortical 0-1 Hz fMRI signal appears in sleep.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkh9JQXsAgTVoB.jpg"}]}, {"id": "1241061134762991617", "user": "lauradata", "date": "2020-03-20T17:56:46+00:00", "text": "6. Next \u2013 what happens to CSF flow during sleep? Sleep is important for brain health but its effects on CSF dynamics were not clear. We used fast fMRI to measure upwards inflow of CSF during NREM sleep \u2013 mostly stage N2: https://t.co/aj8AFwzU0f\n  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkiat8WkAYYhYu.png"}]}, {"id": "1241061524782923777", "user": "lauradata", "date": "2020-03-20T17:58:19+00:00", "text": "7. We found that during N2 sleep, large waves of CSF flow appear, far larger than the typical breathing-locked CSF waves during wakefulness.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkix6QWAAMqz7L.jpg"}]}, {"id": "1241062116817342469", "user": "lauradata", "date": "2020-03-20T18:00:41+00:00", "text": "8. These CSF waves are time-locked to neural slow waves: first EEG slow waves, then fMRI BOLD waves and CSF flow. This coupling suggests hemodynamics (neurovascular coupling, vasomotion, autonomic) as a possible mechanism by which neural activity could modulate CSF flow. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkjTebWAAAU-Jo.jpg"}]}, {"id": "1241062228322811909", "user": "lauradata", "date": "2020-03-20T18:01:07+00:00", "text": "9. So, fast fMRI reveals lots of new info in fMRI signals! High frequency oscillations and CSF flow during sleep. We also need to be careful to account for physiological components of these signals \u2013 e.g.  https://t.co/mymEY3TBIQ\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkjgqrWsAITFzY.jpg"}]}, {"id": "1241062612193902595", "user": "lauradata", "date": "2020-03-20T18:02:39+00:00", "text": "10. These studies use fast fMRI to reveal dynamic network-wide fMRI 0-1 Hz signals and CSF waves during N1/N2 sleep. Our ongoing work is developing joint analyses for EEG-fMRI, dissecting thalamic contributions to sleep dynamics, and studying other sleep stages. Thanks #OHBMx !", "media": []}]}, {"date": "2020-03-20T18:15:07+00:00", "text": "#OHBMx-55 \u2733 #talk\n\nNoam Peled @pelednoam*, O Felsenstein @ohadfel, E Hahn @HahnEmilyM, A Rockhill @rock_on_the_run, L Folsom @lyndefolsom, T Gholipour @tahagholipour, D Dougherty @DarinDougherty, A Paulk @ACPaulk, S Cash, A Widge @AlikWidge, M H\u00e4m\u00e4l\u00e4inen @mshamalai,   S Stufflebeam \n\n*Martinos Center @ MGH\n\n\u25b6 Multi-Modal Analysis and Visualization Tool (MMVT)\n\n#Anatomy #Connectivity #Dynamics #Tools", "media": [], "ids": ["1241065751991267329", "1241065832563777539"], "thread": [{"id": "1241066149162467335", "user": "pelednoam", "date": "2020-03-20T18:16:42+00:00", "text": "1. The Multi-Modal Analysis and Visualization Tool (https://t.co/IGyeG3xYuc) is designed for researchers wishing to better understand their neuroimaging functional and anatomical data through simultaneous visualization of these existing imaging modalities.\n#OHBMx", "media": []}, {"id": "1241066573714071552", "user": "pelednoam", "date": "2020-03-20T18:18:23+00:00", "text": "2. Main MMVT GUI. (a) A 3D brain view, (b) Color-bar that being updated automatically according to the activity being plotted, (c) Slices viewer for MRI (T1, T2, and FLAIR) and CT, (d) Time-domain and frequency-domain graphs, and (e) MMVT panels\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkndRyWsAI1nQi.jpg"}]}, {"id": "1241067567780974593", "user": "pelednoam", "date": "2020-03-20T18:22:20+00:00", "text": "3. In one case study, we compared resting-state connectivity between fMRI and MEG. (a) fMRI (right dmPFC, in red) and MEG (left precuneus in blue). (b) and (c) MEG and fMRI (respectively) main hubs connectivity on a flat brain.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkoDNjWAAAW5Bc.jpg"}]}, {"id": "1241068385603125252", "user": "pelednoam", "date": "2020-03-20T18:25:35+00:00", "text": "4. Electrodes identification using MRI and CT: Using a semi-automatic algorithm, we identify the electrodes' contacts in co-registered CT and MRI and group them to reconstruct the depth electrodes. \n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkpFhaXkAQCbg7.jpg"}]}, {"id": "1241068855264530432", "user": "pelednoam", "date": "2020-03-20T18:27:27+00:00", "text": "5. fMRI analysis VS MEG gamma power: (a) Correlation in the left superiorfrontal between the MEG high-gamma Welch's t-test results in red-yellow colors, and the fMRI contrast map in blue contours. (b) Welch's t-test MEG results over frequencies, where the 78H is selected.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkpco7WsAASV7i.png"}]}, {"id": "1241069123238584322", "user": "pelednoam", "date": "2020-03-20T18:28:31+00:00", "text": "6. MMVT can visualize TMS-evoked EEG potentials in real-time to precisely navigate to the optimal stimulation position and orientation, guided by individual brain morphology. \n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkpxzJWoAM4on_.jpg"}]}]}, {"date": "2020-03-20T18:30:05+00:00", "text": "#OHBMx-56 \u2733 #talk\n\nVince Calhoun @vdcalhoun*, A Iraji @AIraji, R Miller @robynlm\n\n*TReNDS Center\n\n\u25b6 Space: A Missing Piece of the Dynamic Puzzle\n\n#Connectivity #Disorders #Dynamics #Methods", "media": [], "ids": ["1241069515225456640"], "thread": [{"id": "1241069656347041792", "user": "vdcalhoun", "date": "2020-03-20T18:30:38+00:00", "text": "1/6 Space: a Missing Piece of the Dynamic Puzzle: Brain dynamic functional connectivity (dFC) can be temporal, spatial, spatiotemporal. Temporal \u201conly\u201d methods ignore important int-er/ra subject spatial variation and spatially dynamic features (https://t.co/18oW8Va5Zo). #OHBMx", "media": []}, {"id": "1241069988246503424", "user": "vdcalhoun", "date": "2020-03-20T18:31:57+00:00", "text": "2/6 Spatially dynamic is variation in the spatial distribution of a source over time. Temporal dynamics focus on variations in the temporal patterns of sources. An example of temporal, spatial, and spatiotemporal dynamics if the brain has only two functional sources. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkqjpyUUAAYUCv.png"}]}, {"id": "1241070556696293376", "user": "vdcalhoun", "date": "2020-03-20T18:34:13+00:00", "text": "3/6 The same anatomical location may not represent the same source across time. Fixed spatial nodes are not suitable for the study of the spatial dynamics as they assume a priori that the sources are fixed and can produce misleading results. Data driven nodes are needed. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkq6kXU8AAiGOW.png"}]}, {"id": "1241070642469855232", "user": "AIraji", "date": "2020-03-20T18:34:33+00:00", "text": "3a) Each node must represent the same functional source across time; otherwise, the functional connectivity between two nodes may represent relationships between different sources across time.", "media": []}, {"id": "1241070724443332608", "user": "AIraji", "date": "2020-03-20T18:34:53+00:00", "text": "3b) The validity of graph-based models depends on the extent to which nodes and edges represent the true underlying functional sources.", "media": []}, {"id": "1241070753723736068", "user": "AIraji", "date": "2020-03-20T18:35:00+00:00", "text": "3c) We need to incorporate spatially varying nodes in graph-based and connectomic analyses. This also enables us to leverage the advanced mathematical tools from network science and graph theory to study the spatially dynamic properties of the brain.", "media": []}, {"id": "1241071209875247104", "user": "vdcalhoun", "date": "2020-03-20T18:36:49+00:00", "text": "4/6 We can investigate moment-to-moment spatial reconfiguration of brain networks at the finest observable scale (voxels) via a spatially fluid chronnectome model. Brain networks evolve spatially and transiently merge and separate from one another (https://t.co/8qsIIyijH1) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkroImU8AAICkv.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETkrrK4UcAAB-KA.jpg"}]}, {"id": "1241071583390625792", "user": "vdcalhoun", "date": "2020-03-20T18:38:18+00:00", "text": "5/6 We can also use hierarchical models of brain function...each level represents different spatial scales and different dynamic information. Functional domains (FDs) evolve spatially, including changes in a region's association to a given FD https://t.co/4BlYUGOcSh #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkr4KzU8AAwwB9.jpg"}]}, {"id": "1241071913092308992", "user": "AIraji", "date": "2020-03-20T18:39:36+00:00", "text": "5a) In this framework, we construct the brain function as a hierarchical structure based on functional homogeneity. Different levels of the hierarchy represent different spatial scales and contain different dynamic information.", "media": []}, {"id": "1241071946076262402", "user": "AIraji", "date": "2020-03-20T18:39:44+00:00", "text": "5b) A region's association with a given functional domain (FD) can change over time from strong association to complete dissociation.", "media": []}, {"id": "1241071973037273088", "user": "vdcalhoun", "date": "2020-03-20T18:39:51+00:00", "text": "6/6 Allowing for spatial dynamics explains inconsistencies in spatially static studies. It provides new metrics to study the brain and its alterations across disorders. We need dynamic models that can capture the complexity of brain function changes in space AND time. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETksIe_UcAE-bD2.jpg"}]}, {"id": "1241071982851944449", "user": "AIraji", "date": "2020-03-20T18:39:53+00:00", "text": "5c) FDs such as the frontoparietal and attention which are engaged in higher cognitive processing have a higher level of dynamism and spatial variations than those involved in primary processing. Similar findings were observed in multi-task fMRI and temporal dynamics.", "media": []}, {"id": "1241072161877442560", "user": "AIraji", "date": "2020-03-20T18:40:36+00:00", "text": "6a) e.g., different spatial patterns of the default mode network reported in previous static functional connectivity studies appear at different moments of time, highlighting inconsistencies due to methods that do not account for spatial dynamics", "media": []}, {"id": "1241072217481285632", "user": "AIraji", "date": "2020-03-20T18:40:49+00:00", "text": "Individualized, dynamic functional parcellations? https://t.co/ik1WmOqLSL", "media": []}, {"id": "1241072252189196288", "user": "AIraji", "date": "2020-03-20T18:40:57+00:00", "text": "6c) We need to develop approaches that directly utilize the spatial properties, such as distance or direction, when identifying dynamic patterns, or they can evaluate the spatially dynamic behaviors using spatial statistics.", "media": []}, {"id": "1241072295025594368", "user": "AIraji", "date": "2020-03-20T18:41:07+00:00", "text": "6d) Also, structurally-informed dynamic modeling approaches that leverage the spatiostructural information can provide new insights into brain dynamics.", "media": []}]}, {"date": "2020-03-20T18:45:08+00:00", "text": "#OHBMx-57 \u2733 #talk\n\nGael Varoquaux @GaelVaroquaux*, K Dadi, A Mensch @arthurmensch, B Thirion @BertrandThirion, A Machlouzarides-Shalit @Antonia_Machlou, C Gorgolewski @chrisgorgo, D Wassermann @demwassermann\n\n*Inria\n \n\u25b6  DiFuMo: Dictionary of Functional Modes for brain imaging\n\n#Connectivity #Methods #Modeling", "media": [], "ids": ["1241073304116609024", "1241073399474118656"], "thread": [{"id": "1241073937280315394", "user": "GaelVaroquaux", "date": "2020-03-20T18:47:39+00:00", "text": "1/ Growing size of population-imaging cohorts promises insights on neural basis of personality, risk-factors of mental health\n\nData is huge: 4Tb for HCP, much more for UK Biobank.\n\nTo alleviate computing cost, image-derived phenotypes are extracted on brain parcelations\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkuJdtWkAExbHk.jpg"}]}, {"id": "1241074144130805761", "user": "GaelVaroquaux", "date": "2020-03-20T18:48:28+00:00", "text": "2/ We optimize high-dimensional dictionaries of functional modes (DiFuMo) for signal extraction across 27 studies, to give a basis for optimal signal extraction.\nhttps://t.co/ara9dQmFEY\n\nFrom 64 to 1024 brain-wide probabilistic functional modes\n#OHBMx \ud83d\ude03", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkuWGDWoAE77rL.png"}]}, {"id": "1241074322548146177", "user": "GaelVaroquaux", "date": "2020-03-20T18:49:11+00:00", "text": "3/ We show that their extracted signal leads to better statistical analysis of fMRI for decoding, standard analysis and functional connectivity biomarkers, sometimes better than voxel level \ud83d\ude0e.\n\nHigh dimensionality is key for all but functional connectivity.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkugSPXQAAayEJ.jpg"}]}, {"id": "1241074526462578688", "user": "GaelVaroquaux", "date": "2020-03-20T18:49:59+00:00", "text": "4/ They also give excellent summary of images across studies, eg for meta-analysis.\n\n1024 components are sufficient to capture 80% of the variance in functional images, greatly facilitating data handling, giving orders of magnitude decrease in computing time and storage\ud83c\udfc3\u200d\u2642\ufe0f.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkurw_XgAITb7a.jpg"}]}, {"id": "1241074713247584257", "user": "GaelVaroquaux", "date": "2020-03-20T18:50:44+00:00", "text": "5/ Modes are optimized from the data: eg the choice of splitting the putamen across hemispheres or in the anterio-posterior direction is data-driven.\n\nFor this, we used massive dictionary-learning on 2.4TB of fMRI data.\n\nPreprint: https://t.co/7GT654VUpV\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETku1xVWkAAf8xx.png"}]}, {"id": "1241075070950309888", "user": "GaelVaroquaux", "date": "2020-03-20T18:52:09+00:00", "text": "6/ We have shared the atlases with anatomical labeling of each mode (!)\neg https://t.co/Gc55GGOOu7\n\nWe hope that they will define standard IDPs, to facilitate analysis of very large population, while improving sharing, communication, and statistical power\ud83e\udd1f.\n\nStay safe\ud83c\udf0f\ud83c\udf0d\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkvFBBXgAYYABO.jpg"}]}]}, {"date": "2020-03-20T19:00:08+00:00", "text": "#OHBMx-58 \u2733 #talk\n\nReinder Vos de Wael @reindervosdwael*, O Benkarim @oualid_benkarim, C Paquola @CaseyPaquola, S Lariviere @saratheriver, J Royer @J_Royer_, S Tavakol, T Xu @TingsterX, S-J Hong @hong_seok_jun, G Langs @georg, S Valk @sofievalk, B Misic @misicbata,   M Milham , D Margulies @neuro_conn, J Smallwood , B Bernhardt \n\n*McGill University\n\n\u25b6 BrainSpace, the Gradient Connectivity Toolbox\n\n#Connectivity #Tools", "media": [], "ids": ["1241077078096740352", "1241077191401668608"], "thread": [{"id": "1241077873227816964", "user": "reindervosdwael", "date": "2020-03-20T19:03:17+00:00", "text": "1/6 Ever wanted to make your own #corticalgradients? BrainSpace, the Python/MATLAB gradient connectivity toolbox, is your friend! In this thread I'll tell you what the toolbox does, and how you can turn your own data into beautiful gradients with only a few lines of code. #OHBMx", "media": []}, {"id": "1241078343979741184", "user": "reindervosdwael", "date": "2020-03-20T19:05:09+00:00", "text": "2/6 #corticalgradients describe the largest axes of variance in the data. For example, the first gradient of functional connectivity describes a sensorimotor to default mode axis. For more examples see the presentations by @sofievalk, @caseypaquola #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkyJoJX0AIeHJZ.png"}]}, {"id": "1241078583742914560", "user": "reindervosdwael", "date": "2020-03-20T19:06:07+00:00", "text": "3/6 Making your own gradients can be done with only a few lines of code! BrainSpace was built to be simple to use for beginners, whilst also providing the possibility for custom kernels and dimensionality reduction techniques for advanced users. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkyU21X0AQ2t14.jpg"}]}, {"id": "1241078783081435138", "user": "reindervosdwael", "date": "2020-03-20T19:06:54+00:00", "text": "4/6 BrainSpace computes the similarity of each region (=affinity matrix) with a user-specified kernel. A dimensionality reduction technique decomposes this matrix into several gradients (=eigenvectors). These are sorted by their eigenvalues to determine their order. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkyjiJXsAAa7dt.jpg"}]}, {"id": "1241079016687374338", "user": "reindervosdwael", "date": "2020-03-20T19:07:50+00:00", "text": "5/6 Statistical tests on MRI data may be biased by spatial autocorrelation in the data. We provide null models for testing significant associations in the presence of spatial autocorrelation. Currently, BrainSpace includes spin tests and Moran spectral randomization. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETkyvm2WsAsnUpv.jpg"}]}, {"id": "1241079180240060416", "user": "reindervosdwael", "date": "2020-03-20T19:08:29+00:00", "text": "6/6 Install BrainSpace with `pip install brainspace` or download it from https://t.co/gKGLNOqiWS.\n\nDocumentation and tutorials available at: https://t.co/R32W9GnUh7. \n\nSee also our publication: https://t.co/jvgeTzLHIV #OHBMx", "media": []}]}, {"date": "2020-03-20T19:15:04+00:00", "text": "#OHBMx-59 \u2733 #talk\n\nCasey Paquola @CaseyPaquola*, O Benkarim @oualid_benkarim, J DeKraker @jordandekraker, N Bernasconi, A Razi @adeelrazi, A Khan @neuroak, B Bernhardt @BorisBernhardt\n\n*Montreal Neurological Institute\n \n\u25b6 Cortical confluence\n\n#Anatomy #Connectivity #Modeling #Networks", "media": [], "ids": ["1241080838202241028", "1241080936667774976"], "thread": [{"id": "1241081232236204034", "user": "CaseyPaquola", "date": "2020-03-20T19:16:38+00:00", "text": "1\u20e3 Hi \ud83d\ude0a the infolding of the cortex in the medial temporal lobe that forms the hippocampus is evocative - horn of a god, sea monster \u2013 and it also holds the most profound transition in the cellular infrastructure of the cortex, from 6layer isocortex to 3layer allocortex #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk0r6eXsAEGpaJ.jpg"}]}, {"id": "1241081600638681090", "user": "CaseyPaquola", "date": "2020-03-20T19:18:06+00:00", "text": "2\u20e3The hippocampus is often separated from the rest of the cortex, so first we developed a model of the continuous cortical surface in the medial temporal lobe, then mapped a geometric axis running from the deepest aspect of the allocortical hippocampus out to the isocortex #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/ext_tw_video/1241081447588528130/pu/vid/1204x492/ies-1lPLBQU7zOJv.mp4?tag=10", "content-type": "video/mp4"}]}, {"id": "1241081989077315595", "user": "CaseyPaquola", "date": "2020-03-20T19:19:39+00:00", "text": "3\u20e3Combining microstructure profiles and manifold learning, we found that the geometrically defined axis from iso-to-allocortex closely aligns with the principle axis of cytoarchitectural differentiation - i.e. infolding co-occurs with major changes in cellular composition #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk1RBpWkAUAypX.jpg"}]}, {"id": "1241082319248723969", "user": "CaseyPaquola", "date": "2020-03-20T19:20:57+00:00", "text": "4\u20e3After transformation from histological to MNI space, we estimated the direction of information flow using dynamic causal modelling of rs-fMRI. Effective connectivity showed a flow from the iso-to-allocortex, as well as dense interconnectivity within each cortical type #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk1vhkWAAQcGjH.jpg"}]}, {"id": "1241082695960199172", "user": "CaseyPaquola", "date": "2020-03-20T19:22:27+00:00", "text": "5\u20e3Geometric axes of the cortical confluence show clear preferences in rsFC to the rest of the isocortex too, which is especially clear in relation to large-scale functional #gradients\n\nIso-to-allocortex ~ task positive-intrinsic\nPosterior-anterior ~ sensory-transmodal\n\n\ud83c\udf08#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk2EZOWAAUgV5G.jpg"}]}, {"id": "1241083154439581696", "user": "CaseyPaquola", "date": "2020-03-20T19:24:16+00:00", "text": "6\u20e3This cortical confluence really highlights the importance of geometry and cytoarchitecture for understanding brain organisation and function \nSpecial thanks to   and @jordandekraker   for the histological work, and  for rs-fmri, and to #OHBMx!", "media": []}]}, {"date": "2020-03-20T19:30:13+00:00", "text": "#OHBMx-60 \u2733 #talk\n\nAki Nikolaidis @AkiNikolaidis*, A Solon Heinsfeld @AnibalSolon, T Xu @TingsterX, P Bellec @Pierre_bellec, J Vogelstein @neuro_data, M Milham @MichaelMilham\n\n*Child Mind Insititute\n \n\u25b6 Bagging improves reproducibility of functional parcellation\n\n#Applications #Connectivity #Methods #Networks", "media": [], "ids": ["1241084651378298886", "1241084793011437568"], "thread": [{"id": "1241084826247020545", "user": "AkiNikolaidis", "date": "2020-03-20T19:30:55+00:00", "text": "Thank you!! So excited to get to participate in OHBMx and so impressed by this coordinated global effort!", "media": []}, {"id": "1241084864918482944", "user": "AkiNikolaidis", "date": "2020-03-20T19:31:04+00:00", "text": "This study is about how a resampling approach called bagging can be used to improve the reproducibility and reliability of mapping the brain through functional parcellations. #OHBMx \n\nhttps://t.co/jTJ02hVi3K", "media": []}, {"id": "1241084986918199296", "user": "AkiNikolaidis", "date": "2020-03-20T19:31:33+00:00", "text": "1. We studied the reproducibility of group and individual level functional parcellations in two datasets: HNU1 ( n = 30; 10 scans x 10 mins per scan), and GSP ( n = 300; 2 scans x 6 mins per scan). All groups were age and gender matched young adults. #OHBMx", "media": []}, {"id": "1241085062881280000", "user": "AkiNikolaidis", "date": "2020-03-20T19:31:51+00:00", "text": "2. Cortical and subcortical functional parcellations were created in each of these datasets at group and individual levels. We compared the stability of the individual and group parcellations with and without bagging. #OHBMx", "media": []}, {"id": "1241085143617421312", "user": "AkiNikolaidis", "date": "2020-03-20T19:32:11+00:00", "text": "3. We found that: compared to standard clustering approaches, bagging-enhanced parcellations have much higher between scan, between session, between sample, and between study reproducibility! And this effect replicates across scanners, sites, clustering algorithms etc. #OHBMx", "media": []}, {"id": "1241085303940452354", "user": "AkiNikolaidis", "date": "2020-03-20T19:32:49+00:00", "text": "4. Bagging reduces the data req. for reproducible parcellations. I.e. if you have 6 minutes of scan time, bagging can beat reproducibility of non-bagged with 12 minutes. The same holds for 10, 15, 20 minutes of data. Key for most studies with <30mins+ of data #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk4fzDUUAAFkY2.jpg"}]}, {"id": "1241085441463349248", "user": "AkiNikolaidis", "date": "2020-03-20T19:33:22+00:00", "text": "5. We see the largest effects of bagging on between session and between scan reproducibility, which is absolutely key for any studies looking at repeated measures with imaging based outcomes, like intervention studies or longitudinal developmental studies like ABCD #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk4mELUcAAiLUq.jpg"}]}, {"id": "1241085543015829504", "user": "AkiNikolaidis", "date": "2020-03-20T19:33:46+00:00", "text": "6. One surprising effect was that we saw group parcellations became more representative of the individuals that make up that group. In other words, the individual level parcellations don\u2019t vary as much from a group level parcellation created from the same individuals. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk4tVsUwAA0S-0.jpg"}]}]}, {"date": "2020-03-20T19:45:03+00:00", "text": "#OHBMx-61 \u2733 #talk\n\nVaibhav Tripathi @vaibhav_3pathi*\n\n*Boston University\n\n\u25b6 Gender Differences in Cortical fMRI Activity while Movie Watching\n\n#Connectivity #Networks", "media": [], "ids": ["1241088384833118210"], "thread": [{"id": "1241088556170391558", "user": "vaibhav_3pathi", "date": "2020-03-20T19:45:44+00:00", "text": "1/6 Our brain is designed to operate in the natural world. Recent work in fMRI of naturalistic stimuli has opened up amazing avenues of research. In this talk we look at an obvious question: do men and women process movies differently? \ud83d\udc66\ud83d\udc69\u200d\ud83e\uddb0\ud83e\udde0\ud83d\udcbb #OHBMx", "media": []}, {"id": "1241088906445180928", "user": "vaibhav_3pathi", "date": "2020-03-20T19:47:08+00:00", "text": "2/6 We used the Human Connectome Project(HCP) 7T dataset with a subset of 134 subjects(67 females). Each subject underwent 4 runs of 15 minutes of the audiovisual stimulus. Each run had 3 clips either from Hollywood movies or Non-Hollywood Indie movies in English language. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk7sDhXkAE3zZg.jpg"}]}, {"id": "1241089555639525376", "user": "vaibhav_3pathi", "date": "2020-03-20T19:49:43+00:00", "text": "3/6 Functional Connectivity(FC) during movie watching was computed with nodes(360 parcels or 22 sections) defined in HCP-MMP atlas. T-test calculated differences for each cell in the FC matrix. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk8WMOXsAI3OB9.jpg"}]}, {"id": "1241090202065567744", "user": "vaibhav_3pathi", "date": "2020-03-20T19:52:17+00:00", "text": "4/6 Functional connectivity was compared across the different clips for both the genders. One way ANOVA results are computed across FC of each clip individually. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk8uhpWAAEF5FF.jpg"}]}, {"id": "1241091021007699969", "user": "vaibhav_3pathi", "date": "2020-03-20T19:55:32+00:00", "text": "5/6 Intersubject correlation(ISC) analysis was performed for the parcel averaged time-series signals. Parcels in the right frontal lateral cortex(44, AVI, 6r, IFJa, PEF, IFSp) and left inferior temporal regions(VVC, FFC) are more correlated in \ud83d\udc66(r=0.48) than \ud83d\udc69\u200d\ud83e\uddb0(r=0.35). #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETk9aGrXgAAXZW4.mp4", "content-type": "video/mp4"}]}, {"id": "1241091549192163334", "user": "vaibhav_3pathi", "date": "2020-03-20T19:57:38+00:00", "text": "6/6 Gender differences have been reported in anatomical, functional and connectivity. We expanded this to movie processing in the \ud83e\udde0. PCC, insula, dlPFC, mid-cingulate and mPFC are regions which differ in FC/ISC correlations across genders for naturalistic movies. #OHBMx", "media": []}]}, {"date": "2020-03-20T20:00:02+00:00", "text": "#OHBMx-62 \u2733 #talk\n\nRussell Poldrack @russpoldrack*, R Botvinik-Nezer @rotembot, T Glatard @TristanGlatard, T Nichols @ten_photos, T Schonberg @tschonberg\n\n*Stanford University\n\n\u25b6 Fully reproducible data analysis in the NARPS study\n\n#Methods #Tools", "media": [], "ids": ["1241092154988138497"], "thread": [{"id": "1241092479731896320", "user": "russpoldrack", "date": "2020-03-20T20:01:20+00:00", "text": "1) In the NARPS Study #ohbmx-17 we asked how analysis results from a single dataset would vary across groups. 70 analysis teams submitted hypothesis tests and thresholded/unthresholded maps. Original data on  https://t.co/PXNCd9CQEf #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk-7z2UwAADR19.jpg"}]}, {"id": "1241092828458962944", "user": "russpoldrack", "date": "2020-03-20T20:02:43+00:00", "text": "2) Here we will describe a completely reproducible end-to-end analysis pipeline to analyze those results from the participating groups.  The goal was to allow any researcher to fully reproduce the results reported in the paper.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk_VaEUMAA6-sd.jpg"}]}, {"id": "1241093158353551360", "user": "russpoldrack", "date": "2020-03-20T20:04:02+00:00", "text": "3) The analysis results submitted by 70 teams to  were organized and shared via   https://t.co/gXAHsBFqX0.  Code was shared via Github https://t.co/H7duHs1nJv  and versioned releases shared via Zenodo.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk_nNtUEAAX3Dp.jpg"}]}, {"id": "1241093479926620160", "user": "russpoldrack", "date": "2020-03-20T20:05:18+00:00", "text": "4) All processing was implemented within a  container, with all software versions fixed.  Python versions specified directly in Dockerfile, R versions using checkpoint package https://t.co/PrpM3KUfZi #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETk_xvAUUAAxi6S.jpg"}]}, {"id": "1241093706578407424", "user": "russpoldrack", "date": "2020-03-20T20:06:12+00:00", "text": "5) Full processing stream was run automatically using  continuous integration (upon any push of new code to Github). Results exported and shared via  https://t.co/myg9NSC4kj #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlABt_U8AA7uZr.jpg"}]}, {"id": "1241093898253897728", "user": "russpoldrack", "date": "2020-03-20T20:06:58+00:00", "text": "6) Many thanks to coauthors @rotembot @TristanGlatard @ten_photos @tschonberg and entire NARPS group!  See https://t.co/WFFVUpr1EV and for more details and https://t.co/3EMz0QD20T for preprint #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlAN_YUEAEhGWN.jpg"}]}]}, {"date": "2020-03-20T20:15:06+00:00", "text": "#OHBMx-63 \u2733 #talk\n\nMengjia Xu @xx_mengjia*, D L Sanz, P Garces, F Maestu, Q Li, D Pantazis @dimitrpantazis\n\n*McGovern Institute for Brain Research, MIT\n \n\u25b6 MEG brain network Gaussian embeddings for predicting Alzheimer\u2019s disease progression\n\n#Applications #Connectivity #Methods #Networks", "media": [], "ids": ["1241095945380601863", "1241096019565305863"], "thread": [{"id": "1241189688007852035", "user": "xx_mengjia", "date": "2020-03-21T02:27:36+00:00", "text": "1/6 A potential biomarker for Alzheimer\u2019s disease (AD) is the disruption of functional brain networks measured by MEG. Here we developed a multiple graph Gaussian embedding method with uncertainty quantification for predicting AD progression using MEG data.  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmXaX8XYAEopCD.jpg"}]}, {"id": "1241189832195481602", "user": "xx_mengjia", "date": "2020-03-21T02:28:10+00:00", "text": "2/6 We investigate if subtle alterations in MEG brain networks can predict AD progression. Measurements is from Madrid cohort database with stable/progressive MCI and control subjects. 5 mins eye-closed resting state MEG data was collected using a 306-channel MEG system. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmXkCvXgAIC083.jpg"}]}, {"id": "1241189986390757377", "user": "xx_mengjia", "date": "2020-03-21T02:28:47+00:00", "text": "3/6 We computed source-level MEG brain networks using DKT atlas, and built a stochastic embedding method to learn informative embeddings with uncertainty quantification. We applied the latent embeddings to predict AD progression and quantify group-wise changes. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmXtD2WkAEyxpY.jpg"}]}, {"id": "1241190103663480832", "user": "xx_mengjia", "date": "2020-03-21T02:29:15+00:00", "text": "4/6 With 3D stochastic encoder architecture, we transform high-dimensional MEG brain networks into a latent node-wise stochastic embedding space (i.e., multivariate normal distributions). The optimal embedding dimension obtained by performing link prediction experiments. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmX0GVWAAAqmRJ.jpg"}]}, {"id": "1241190209288536074", "user": "xx_mengjia", "date": "2020-03-21T02:29:40+00:00", "text": "5/6 The transformed MEG brain network Gaussian embedding signature is employed for subsequent important tasks of AD progression, and for effective discrimination of stable and progressive MCI. This can provide potential insights into AD phenotyping. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmX6RCWoAImvOk.jpg"}]}, {"id": "1241190291530530816", "user": "xx_mengjia", "date": "2020-03-21T02:30:00+00:00", "text": "6/6 \u201cgroup-wise\u201d changing regions in preclinical stages of AD are detected using a statistic permutation approach and FDR correction with the probabilistic embeddings. Visualization results indicated that most significant regions fall into the temporal and frontal lobes. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmX_BtWoAE7WZH.jpg"}]}]}, {"date": "2020-03-20T20:30:02+00:00", "text": "#OHBMx-64 \u2733 #talk\n\nCaroline Magnain @CMagnain*, B Edlow @ComaRecoverylab, J Augustinack, A van der Kouwe, C Jaimes, L-M Terrier @Harvey_Cushing2, H Freeman, E Boyd @emmaboyd23, L Morgan, M Fogarty @Its_MorganF, L Ferraz da Silva, E Haas, R Haynes, B Fischl @BruceFischl,   L Zollei @lzollei\n\n*Martinos Center MGH/HMS\n\n\u25b6 High Resolution Post-mortem Imaging of the Infant Brainstem Arousal Network\n\n#Anatomy #Consciousness #Developmental #Methods", "media": [], "ids": ["1241099705125347329", "1241099785014296576"], "thread": [{"id": "1241100077000687616", "user": "CMagnain", "date": "2020-03-20T20:31:31+00:00", "text": "(1) Sudden infant death syndrome (SIDS) is the leading cause of postneonatal infant mortality in industrialized nations [Goldstein et al, 2016 Pediatrics 137:1-10] #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlF3s_X0AMP556.jpg"}]}, {"id": "1241100724139851777", "user": "CMagnain", "date": "2020-03-20T20:34:05+00:00", "text": "(2) A leading hypothesis: Arousal Deficit Hypothesis -- (a subset of) SIDS is caused by a failure in arousal to a life-threatening, sleep-related stressor during the critical developmental period of the first year of life. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlGZhDXgAIgZ7x.png"}]}, {"id": "1241101288248610818", "user": "CMagnain", "date": "2020-03-20T20:36:20+00:00", "text": "(3) Our goal is to obtain an accurate anatomical representation of the infant brainstem arousal network through complementary ex vivo imaging techniques and their analysis. https://t.co/t55KHP5w8l #OHBMx", "media": []}, {"id": "1241101790256541697", "user": "CMagnain", "date": "2020-03-20T20:38:20+00:00", "text": "(4) We hypothesize that high-resolution MRI provides sufficient anatomical detail to map the connectivity of the normally developing Ascending Arousal Network (AAN). https://t.co/gXX7DTNgG4 #OHBMx", "media": []}, {"id": "1241102479858774016", "user": "CMagnain", "date": "2020-03-20T20:41:04+00:00", "text": "(5) We use Optical Coherence Tomography (OCT), a high resolution and high contrast volumetric imaging technique (no dyes required and minimal distortions) to accurately segment the brainstem and serve as a guide for diffusion analysis. https://t.co/f3JiOeJQmb  #OHBMx", "media": []}, {"id": "1241102990968328194", "user": "CMagnain", "date": "2020-03-20T20:43:06+00:00", "text": "(6) Preliminary results show that combined ex vivo imaging techniques have the potential to accurately visualize the brainstem with its numerous nuclei and tracts and to localize them in the context of the full brain. https://t.co/1QYKM2dhsQ  \nThanks #OHBMx", "media": []}]}, {"date": "2020-03-20T20:45:12+00:00", "text": "#OHBMx-65 \u2733 #talk\n\nChad Rogers @chadsrogers*, M S Jones, S McConkey, B Spehar, K J Van Engen @kj_vanengen, M S Sommers, J E Peelle @jpeelle\n\n*Union College\n\n\u25b6 Spoken words elicit less auditory cortex activity with advancing age\n\n#Developmental #Language #Motor #Sensory", "media": [], "ids": ["1241103520880824323"], "thread": [{"id": "1241104064160706568", "user": "ChadsRogers", "date": "2020-03-20T20:47:22+00:00", "text": "1\ufe0f\u20e3Our ability to hear declines with age, particularly for speech.\ud83d\udde3\ud83d\udc42\ud83e\udde0  \nLess known is the extent to which brain systems that support speech perception change with age.\nw/ @jpeelle, @kj_vanengen, @brentspehar,& non-twittering co-authors. Pre-print: https://t.co/JK8woK3ZGA #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlJjisWsAIPOeb.jpg"}]}, {"id": "1241104392360808452", "user": "ChadsRogers", "date": "2020-03-20T20:48:40+00:00", "text": "2\ufe0f\u20e3We examined brain activity using fMRI in young and older adults while listening to speech in quiet (no background noise, sparse sampling). Listeners did repetition and active listening tasks. Hearing data were available on a subset of participants. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlJ1UJWsAYcE94.jpg"}]}, {"id": "1241104761467830276", "user": "ChadsRogers", "date": "2020-03-20T20:50:08+00:00", "text": "3\ufe0f\u20e3We looked in probabilistically-defined auditory cortex and found:\n\n-Similar activation in young and older adults for our noise condition\n\n-Older adults showed reliably less activation than young when listening to speech.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlKIvxX0AoDBqF.jpg"}]}, {"id": "1241105122542915589", "user": "ChadsRogers", "date": "2020-03-20T20:51:34+00:00", "text": "4\ufe0f\u20e3Older adults' activity didn't correlate with accuracy or hearing ability (or movement, not shown).\n\n #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlKgAPWAAMO9J4.jpg"}]}, {"id": "1241105527825928199", "user": "ChadsRogers", "date": "2020-03-20T20:53:11+00:00", "text": "5\ufe0f\u20e3Whole-brain analyses on both tasks revealed similar patterns of activation for young and older adults. Young adults showed more activation in superior temporal cortex near auditory cortex. \n\nNo sig clusters for older > young.\n\n #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlKtbVXgAEQpS9.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETlKvJVWoAAUWsg.jpg"}]}, {"id": "1241105780843175936", "user": "ChadsRogers", "date": "2020-03-20T20:54:11+00:00", "text": "6\ufe0f\u20e3SUMMARY: Older vs. young adults\u2019 activation during spoken word recognition largely the same, but with less activity in auditory cortex in older adults.\n\nSimilar results for listen and repeat tasks\n\n#OHBMx", "media": []}]}, {"date": "2020-03-20T21:00:06+00:00", "text": "#OHBMx-66 \u2733 #talk\n\nBenjamin Balas @bjbalas*, A Saville\n\n*North Dakota State University\n\n\u25b6 Developmental Changes in ERP responses to Natural Texture Statistics\n\n#Developmental #sensory", "media": [], "ids": ["1241107268441079809"], "thread": [{"id": "1241107535828070400", "user": "bjbalas", "date": "2020-03-20T21:01:09+00:00", "text": "1) Natural images have lawful properties that the adult visual system is tuned to. For example, adults show increased sensitivity for patterns with amplitude spectrum fall-off coefficients close to those observed in natural scenes. #OHBMx", "media": []}, {"id": "1241108050116739072", "user": "bjbalas", "date": "2020-03-20T21:03:12+00:00", "text": "2) Kid\u2019s visual systems continue to develop to adult-like levels of contrast sensitivity, spatial frequency discrimination, and contour integration during middle childhood. While some  low-level mechanisms are mature early in childhood, others are not adult-like until 10. #OHBMx", "media": []}, {"id": "1241108929523965953", "user": "bjbalas", "date": "2020-03-20T21:06:42+00:00", "text": "3) We investigated whether 5-10 y.o. kids were sensitive to deviations from natural image statistics. We examined the response of the P1 and N1 ERP components to two kinds of deviation: contrast negation and the application of texture synthesis models. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlN7KUXYAAXsC0.jpg"}]}, {"id": "1241109363311480832", "user": "bjbalas", "date": "2020-03-20T21:08:25+00:00", "text": "4) These manipulations target different scales of natural image statistics: Contrast negation preserves global layout, while reversing local edge polarity. Texture synthesis matches statistics in small-scale neighborhoods, but disrupts larger scale visual structure. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlOXzDX0AED9Dw.jpg"}]}, {"id": "1241110276356833280", "user": "bjbalas", "date": "2020-03-20T21:12:03+00:00", "text": "5) At the P1 component, we found that only young children (5-7 years) were sensitive to contrast polarity (slower P1 latencies to negative-contrast images). At the N1 component, we observed main effects of synthetic appearance and contrast polarity in all age groups. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlPHVbXkAYmBD-.png"}]}, {"id": "1241111162252595200", "user": "bjbalas", "date": "2020-03-20T21:15:34+00:00", "text": "6) Why do young children show early-stage sensitivity to contrast polarity but older Os don\u2019t? This may reflect the end of a sort of \u201cpre-constancy\u201d version of natural image processing that hasn't yet developed fully invariant recognition. Or it could be something else. \ud83d\ude00 #OHBMx", "media": []}]}, {"date": "2020-03-20T21:16:05+00:00", "text": "#OHBMx-000 \u2733 #break\n\n\u25b6 COFFEE BREAK https://t.co/05KCuHDdE3", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlQGURXkAIsPcO.jpg"}], "ids": ["1241111290380238849"], "thread": []}, {"date": "2020-03-20T21:30:09+00:00", "text": "#OHBMx-67 \u2733 #keynote\n\nLucina Uddin @LucinaUddin*\n\n*University of Miami\n\n\u25b6 KEYNOTE: Towards a universal taxonomy of brain networks", "media": [], "ids": ["1241114834428661762"], "thread": [{"id": "1241115149995474947", "user": "LucinaUddin", "date": "2020-03-20T21:31:25+00:00", "text": "Hi everyone, this is Lucina Uddin from the Brain Connectivity and Cognition Laboratory at the University of Miami (https://t.co/teN1rwxl41). I\u2019m excited to be delivering my first Twitter Keynote for #OHBMequinox #OHBMx !", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlTlPjXgAEYIRP.jpg"}]}, {"id": "1241115874054082560", "user": "LucinaUddin", "date": "2020-03-20T21:34:17+00:00", "text": "2) I\u2019d like to share with you some theoretical work with  and  . We've been thinking about how cognitive and network neuroscience can move towards a universal taxonomy of brain networks. Here they are during better times in Singapore (2018) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlUMhDX0AA3fKg.jpg"}]}, {"id": "1241116609915293697", "user": "LucinaUddin", "date": "2020-03-20T21:37:13+00:00", "text": "3) Why do we need this? Though the idea that the brain is composed of multiple macro-scale functional networks is ubiquitous in cognitive neuroscience, we have yet to reach consensus on terminology, making it difficult to compare findings from different research groups. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlU-bpXgAUiIJ3.mp4", "content-type": "video/mp4"}]}, {"id": "1241117066360500226", "user": "LucinaUddin", "date": "2020-03-20T21:39:02+00:00", "text": "4) What naming conventions, if universally adopted, will provide the most utility and facilitate communication amongst researchers? Let\u2019s start with something relatively uncontroversial, like networks involved in visual or motor function. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlVJ7nWsAAQOve.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETlVNy_WsBMse1i.jpg"}]}, {"id": "1241117739797938194", "user": "LucinaUddin", "date": "2020-03-20T21:41:42+00:00", "text": "5) We propose that a scheme incorporating anatomical terminology should provide the foundation for a network taxonomy. Instead of naming networks after our favorite cognitive processes, we could build a taxonomy grounded in objective anatomical criteria, such as this #OHBMx.", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlV_ACXYAAXWUP.jpg"}]}, {"id": "1241118444566843397", "user": "LucinaUddin", "date": "2020-03-20T21:44:30+00:00", "text": "6) Of course, there will be controversies. For example, we must consider whether the so-called \u201csalience\u201d, \u201ccingulo-opercular\u201d, and \u201cventral attention\u201d networks constitute distinct entities, or represent versions of the same network. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlWZpQXYAAL6WS.jpg"}]}, {"id": "1241119128645185537", "user": "LucinaUddin", "date": "2020-03-20T21:47:13+00:00", "text": "7) Likewise, what about the so-called \"central executive/cognitive control/multiple demand/frontoparietal network\"? Are all researchers using these terms referring to the same entity? #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlW48nXkAgbWkr.jpg"}]}, {"id": "1241120068869775364", "user": "LucinaUddin", "date": "2020-03-20T21:50:57+00:00", "text": "8) For starters, we might delineate \u201coccipital\u201d, \u201cpericentral\u201d, \u201cdorsal frontoparietal\u201d, \u201clateral frontoparietal\u201d, \u201cmidcingulo-insular\u201d, and \u201cmedial frontoparietal\u201d networks. Corresponding cognitive terms are visual, somatomotor, attention, control, salience, and default. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlYHzqWkAcOxgl.mp4", "content-type": "video/mp4"}]}, {"id": "1241120565483704320", "user": "LucinaUddin", "date": "2020-03-20T21:52:56+00:00", "text": "9) Now that I\u2019ve got everyone worked up, let\u2019s discuss some outstanding issues. What about dynamics? Inter-individual variability? Subcortical brain regions? Network fractionation, subsystems, and hierarchies? #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlYktDWoAAzWpb.mp4", "content-type": "video/mp4"}]}, {"id": "1241121709203304449", "user": "LucinaUddin", "date": "2020-03-20T21:57:29+00:00", "text": "10) And finally, let\u2019s brainstorm some data-driven approaches for answering the question \u201cHow many networks are there?\u201d Anyone interested in being part of an @OHBM \u2018best practices\u2019 working group to further refine this proposed taxonomy? https://t.co/vKKKyT0ssX Thanks all! #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlZm-bWAAA-tkD.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T22:00:05+00:00", "text": "#OHBMx-68 \u2733 #talk\n\nUrsula Tooley @utooley*, A T Park @anneparkt, J A Leonard @julia_a_leonard, D S Bassett @danisbassett, A P Mackey @allymackey\n\n*UPenn\n\n\u25b6 Functional Brain Network Development During Early Childhood\n\n#Connectivity #Developmental #Networks", "media": [], "ids": ["1241122366715047939"], "thread": [{"id": "1241122851136118784", "user": "UTooley", "date": "2020-03-20T22:02:01+00:00", "text": "Thanks for having me!\n\n1. Kids develop rapidly early in childhood, as any parent can tell you. What changes in functional brain network architecture occur during early childhood? #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlapKtXsAAf0P3.jpg"}]}, {"id": "1241123107739500552", "user": "UTooley", "date": "2020-03-20T22:03:02+00:00", "text": "2. We examined the brain networks of a cross-sectional sample of 70 kids ages 4-10, controlling for motion, sex, total amount of data, and average network weight. \n\nMeasures of network segregation are consistently positively associated with age. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETla2_6WsAAKQ5M.jpg"}]}, {"id": "1241123961359958016", "user": "UTooley", "date": "2020-03-20T22:06:26+00:00", "text": "3. Increased segregation could be either within-system connectivity \u2191, or between-system connectivity decreasing \u2193. \n\nWhich is it?\n\nWithin-system age effects are weak: we found pos. effects in visual and ventral attention systems but these did not pass FDR correction. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlbmqAWsAYrwuL.jpg"}]}, {"id": "1241124490643374080", "user": "UTooley", "date": "2020-03-20T22:08:32+00:00", "text": "4. But what *really* drives \u2191 segregation is conn. between the default and attentional systems, which is strongly neg. associated with age.  \n\nThis suggests increasing separation between externally-directed attention and internally-oriented cognition as kids get older. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlcE4rWoAA1bx3.mp4", "content-type": "video/mp4"}]}, {"id": "1241124749398487041", "user": "UTooley", "date": "2020-03-20T22:09:33+00:00", "text": "5. Interestingly, connectivity between the visual and dorsal attention systems was *positively* associated with age.\n\nThis hints at the continued development of top-down attentional processes in early childhood. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlcWJ1XQAkgQuj.jpg"}]}, {"id": "1241125130866229249", "user": "UTooley", "date": "2020-03-20T22:11:04+00:00", "text": "In early childhood, kids\u2019 brain networks are becoming increasingly refined. Future analyses will examine links to cognition, how early experience might shape these brain networks, and longitudinal changes in brain network architecture. #OHBMx \n\nThanks all!", "media": []}]}, {"date": "2020-03-20T22:15:07+00:00", "text": "#OHBMx-69 \u2733 #talk\n\nMatteo Fraschini @matteo_brainnet*, M Demuru\n\n*University of Cagliari\n\n\u25b6 EEG fingerprinting\n\n#Methods #Tools", "media": [], "ids": ["1241126148974825474"], "thread": [{"id": "1241126400788217857", "user": "matteo_brainnet", "date": "2020-03-20T22:16:07+00:00", "text": "Hi everyone.\n\n1. Scalp EEG has been extensively investigated as source of neurophysiological features to be used as biometric system. To date, an impressive number of feature extraction techniques have been exploited as potential fingerprints to identify individuals.\n#OHBMx", "media": []}, {"id": "1241126888023691264", "user": "matteo_brainnet", "date": "2020-03-20T22:18:03+00:00", "text": "2. Most approaches need arbitrary choices: \n- frequency band definition\n- selection of correlation metric for FC\n- threshold to reconstruct a network\nBut, EEG exhibits a 1/f-like PS - an aperiodic component that may be characterized in terms of slope and offset\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETleEajXQAEZStw.jpg"}]}, {"id": "1241127423346905088", "user": "matteo_brainnet", "date": "2020-03-20T22:20:11+00:00", "text": "3. Despite the interest in this approach, it has not been investigated yet how it depicts individual variability in EEG. Here, we quantified the aperiodic component of PS (spectral slope and the offset) and used these features to identify subjects in a large EEG dataset\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlejJDXkAAhT-t.jpg"}]}, {"id": "1241127917373067264", "user": "matteo_brainnet", "date": "2020-03-20T22:22:09+00:00", "text": "4. Here we:\n- applied ADJUST to reduce the effects of artifacts\n- segmented (1m) RS EEGs into 5 non-overlapping epochs\n- extracted aperiodic component features (FOOOF toolbox)\n- extracted periodic component features (the relative power of theta, alpha, beta, and gamma)\n#OHBMx", "media": []}, {"id": "1241128390855360513", "user": "matteo_brainnet", "date": "2020-03-20T22:24:02+00:00", "text": "5. Genuine and impostor similarity scores were computed for each analysis separately. \nThe equal error rate (EER), the point where the false acceptance rate equals the false rejection rate, is reported to summarize the results.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlfd8FWsAAv005.jpg"}]}, {"id": "1241128890132844550", "user": "matteo_brainnet", "date": "2020-03-20T22:26:01+00:00", "text": "6. Summary: the aperiodic component is characterized by strong subject-specific properties and outperforms those by canonical spectral features (band-specific)\nMay a biometric-based approach be of help to compare groups? We think so!\nPreprint https://t.co/ucATZVL6RQ\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlf5Y9XQAADxho.png"}]}]}, {"date": "2020-03-20T22:30:03+00:00", "text": "#OHBMx-70 \u2733 #talk\n\nAmanPreet Badhwar @Aman_Badhwar*, G McFall @mcfall_peggy, S Sapkota, S Black, H Chertkow, S Duchesne @SimonDuchesne, M Masellis, L Li, R Dixon, P Bellec @pierre_bellec\n\n*CRIUGM, University of Montreal\n \n\u25b6 A multiomics approach to biomarkers in Alzheimer\u2019s disease\n\n#Connectivity #Disorders #Memory #Neurology", "media": [], "ids": ["1241129907092107265", "1241129972183511040"], "thread": [{"id": "1241130160675586051", "user": "Aman_Badhwar", "date": "2020-03-20T22:31:04+00:00", "text": "1\nAlzheimer\u2019s disease (AD) is a complex, multifactorial pathology. Aetiological & clinical heterogeneity is increasingly recognized as a common characteristic of AD. This heterogeneity complicates diagnosis, treatment & the design and testing of new drugs. #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlhOvDWoAo5-tX.mp4", "content-type": "video/mp4"}]}, {"id": "1241130775258574848", "user": "Aman_Badhwar", "date": "2020-03-20T22:33:30+00:00", "text": "2\n\u2018Multiomics\u2019 biomarkers can help identify AD patient subgroups with homogeneous pathophysiological signatures. We discuss data reduction analyses that identify complementary disease-relevant perturbations from neuroimaging-based Connectomics, Metabolomics & Genomics. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlhZjyXgA0v90W.png"}]}, {"id": "1241131413174390784", "user": "Aman_Badhwar", "date": "2020-03-20T22:36:02+00:00", "text": "3\nBrain subtypes helps stratify patients based on the similarity of their brain imaging features. Different flavors of data-driven subtyping methods are employed for generation of brain subtypes. As shown using ADNI data, they tend to give similar results (good thing!). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlh-azXQAEuE85.png"}]}, {"id": "1241132004353216522", "user": "Aman_Badhwar", "date": "2020-03-20T22:38:23+00:00", "text": "4\nDiscriminant metabolites identified in metabolomics experiment can be combined into Metabolite Panels to increase discriminative power in AD prediction & progression. Analysis of AD metabolite panels pointed to alterations in amino acid, lipid & nucleic acid metabolism #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlisUXWAAAJJWu.png"}]}, {"id": "1241132985782603780", "user": "Aman_Badhwar", "date": "2020-03-20T22:42:17+00:00", "text": "5\nA popular manageable measure for genomics is Polygenic Risk Scores: constructed using (1) GWAS significant SNPs, (2) nominally associated SNPs based on a specified significance threshold, or (3) combinations of mechanism related SNPs (6 main mechanistic clusters shown).\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETljqFWXkAEnaMz.png"}]}, {"id": "1241133680875245573", "user": "Aman_Badhwar", "date": "2020-03-20T22:45:03+00:00", "text": "6\nIn AD, multiomics biomarkers have the potential to reshape clinical diagnosis & define new \u2018bottom-up\u2019 cohorts based on markers of underlying pathologies. In this slide, we propose a roadmap for parsing heterogeneity in AD using a multiomics approach. THANK YOU!\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlkSh0X0AAZY0_.png"}]}]}, {"date": "2020-03-20T22:45:27+00:00", "text": "#OHBMx-71 \u2733 #talk\n\nJohn Samuelsson @JohnSamuelsson1*, P Sundaram, S Khan @dr_sheraz, M Sereno, M H\u00e4m\u00e4l\u00e4inen @mshamalai\n\n*Massachusetts Institute of Technology\n\n\u25b6 Detectability of Cerebellar Activity with MEG and EEG\n\n#Tools #Cerebellum", "media": [], "ids": ["1241133782096392198"], "thread": [{"id": "1241134139086077953", "user": "JohnSamuelsson1", "date": "2020-03-20T22:46:52+00:00", "text": "(1) The cerebellum contains more than 70 % of all neurons in the brain and has been implicated in diseases such as schizophrenia and PD. \n\nThe cerebellum has, however, remained largely overlooked and its electrophysiology remains poorly characterized in health and disease. #OHBMx", "media": []}, {"id": "1241134139086077953", "user": "JohnSamuelsson1", "date": "2020-03-20T22:46:52+00:00", "text": "(1) The cerebellum contains more than 70 % of all neurons in the brain and has been implicated in diseases such as schizophrenia and PD. \n\nThe cerebellum has, however, remained largely overlooked and its electrophysiology remains poorly characterized in health and disease. #OHBMx", "media": []}, {"id": "1241134513763254272", "user": "JohnSamuelsson1", "date": "2020-03-20T22:48:21+00:00", "text": "(2) Here we investigated if MEG or EEG could be used to detect cerebellar activity by performing simulations using a high-resolution tessellation of the cerebellar cortex constructed from repetitive 9.4 T structural MRI of an ex vivo cerebellum (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETllI5hX0Acy1WL.jpg"}]}, {"id": "1241134513763254272", "user": "JohnSamuelsson1", "date": "2020-03-20T22:48:21+00:00", "text": "(2) Here we investigated if MEG or EEG could be used to detect cerebellar activity by performing simulations using a high-resolution tessellation of the cerebellar cortex constructed from repetitive 9.4 T structural MRI of an ex vivo cerebellum (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETllI5hX0Acy1WL.jpg"}]}, {"id": "1241135250572480512", "user": "JohnSamuelsson1", "date": "2020-03-20T22:51:17+00:00", "text": "(3) Using the detailed cerebellar model, we could see that the cerebellar signals from single dipoles were on average about 30 % weaker than the cortical signals, and cancellation was 20-50 % worse due to the tightly folded cerebellar cortex (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETllfDMXYAEz8kF.jpg"}]}, {"id": "1241135250572480512", "user": "JohnSamuelsson1", "date": "2020-03-20T22:51:17+00:00", "text": "(3) Using the detailed cerebellar model, we could see that the cerebellar signals from single dipoles were on average about 30 % weaker than the cortical signals, and cancellation was 20-50 % worse due to the tightly folded cerebellar cortex (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETllfDMXYAEz8kF.jpg"}]}, {"id": "1241135693256167425", "user": "JohnSamuelsson1", "date": "2020-03-20T22:53:03+00:00", "text": "(4) Combined, these two effects make the net cerebellar signals on average 30 - 60 % weaker than the cortical signals. \n\nBut the variation in these results are quite large as a function of position, so we made M/EEG sensitivity maps of the cerebellum (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlmF0vXQAA_4D2.jpg"}]}, {"id": "1241135693256167425", "user": "JohnSamuelsson1", "date": "2020-03-20T22:53:03+00:00", "text": "(4) Combined, these two effects make the net cerebellar signals on average 30 - 60 % weaker than the cortical signals. \n\nBut the variation in these results are quite large as a function of position, so we made M/EEG sensitivity maps of the cerebellum (image). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlmF0vXQAA_4D2.jpg"}]}, {"id": "1241136199441567744", "user": "JohnSamuelsson1", "date": "2020-03-20T22:55:03+00:00", "text": "(5)We see that MEG and EEG have complementary sensitivity distributions; EEG is sensitive to activity in the anterior lobe and MEG to activity in the posterior lobe. \n\nBoth are optimally sensitive to activity in crus I/lobule VI which activate during working memory tasks. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlmjXdXgAcomad.jpg"}]}, {"id": "1241136199441567744", "user": "JohnSamuelsson1", "date": "2020-03-20T22:55:03+00:00", "text": "(5)We see that MEG and EEG have complementary sensitivity distributions; EEG is sensitive to activity in the anterior lobe and MEG to activity in the posterior lobe. \n\nBoth are optimally sensitive to activity in crus I/lobule VI which activate during working memory tasks. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlmjXdXgAcomad.jpg"}]}, {"id": "1241136850519154691", "user": "JohnSamuelsson1", "date": "2020-03-20T22:57:39+00:00", "text": "(6) To conclude, our simulations suggested that the average cerebellar signal in MEG and EEG is only 30\u201360 % weaker than the cortical signal and that MEG and EEG have highly complementary sensitivity distributions. \nCheck out the full article here\nhttps://t.co/22TXRjtywU \n#OHBMx", "media": []}, {"id": "1241136850519154691", "user": "JohnSamuelsson1", "date": "2020-03-20T22:57:39+00:00", "text": "(6) To conclude, our simulations suggested that the average cerebellar signal in MEG and EEG is only 30\u201360 % weaker than the cortical signal and that MEG and EEG have highly complementary sensitivity distributions. \nCheck out the full article here\nhttps://t.co/22TXRjtywU \n#OHBMx", "media": []}]}, {"date": "2020-03-20T23:00:03+00:00", "text": "#OHBMx-72 \u2733 #talk\n\nGuiomar Niso @GuiomarNiso*, @BIDSstandard, S Appelhoff @stefanappelhoff, F Feingold @franklincrn1, M Ganz @melanieganzben1, C Markiewicz @effigies, R Oostenveld @oostenvr, R Poldrack @russpoldrack, K Whitaker @kirstie_j\n\n*Universidad Politecnica de Madrid\n \n\u25b6 BIDS: a data standard to support the neuroimaging community\n\n#Tools #OpenScience", "media": [], "ids": ["1241137457674047489", "1241137517778395136"], "thread": [{"id": "1241137689568645127", "user": "BIDSstandard", "date": "2020-03-20T23:00:59+00:00", "text": "(1) The Brain Imaging Data Structure (BIDS) is a common standard for organizing, describing and sharing neuroimaging data. https://t.co/40tEijPYro\n\nBIDS is based on simple file formats & folder structures that can readily expand to additional data modalities & applications\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETloEvKWAAAiI0Q.mp4", "content-type": "video/mp4"}]}, {"id": "1241138063436394496", "user": "BIDSstandard", "date": "2020-03-20T23:02:28+00:00", "text": "(2) A global community of 200+ people created BIDS standards for many neuroimaging modalities & more to come:\n\n\ud83e\uddf2 MRI: https://t.co/sDLjS9ylrR\n\ud83e\udd91 MEG: https://t.co/ivUdAy0i3b\n\ud83e\udde0 iEEG: https://t.co/R09OyMDuCa\n\u26a1 EEG: https://t.co/UPLSBNrPoa\n\u2622\ufe0f PET: https://t.co/msdWYggy0b\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETloaa9WAAYp1fw.png"}]}, {"id": "1241138403015561216", "user": "BIDSstandard", "date": "2020-03-20T23:03:49+00:00", "text": "(3) Storing #data & #metadata in the right folders is great for you & future users (eg @OpenNeuroorg).\n\n#BIDSApps are  containers with data processing workflows to help researchers be faster & more #reproducible when running their analyses: https://t.co/MMFkwqhFtm\n\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlopoXXQAElF48.mp4", "content-type": "video/mp4"}]}, {"id": "1241142072159211520", "user": "BIDSstandard", "date": "2020-03-20T23:18:23+00:00", "text": "(4) BIDS is a community-led standard following 3 foundational principles:\n\n\ud83d\ude05 To minimize complexity and facilitate adoption\n\ud83d\uddc3\ufe0f Tackle 80% of the most commonly used neuroimaging data, derivatives, and models \n\ud83c\udf0f Engagement of the global neuroimaging community\n\n#ThatsYou #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlsEmAWsAESFdp.mp4", "content-type": "video/mp4"}]}, {"id": "1241142195186475013", "user": "BIDSstandard", "date": "2020-03-20T23:18:53+00:00", "text": "(5) In 2019 BIDS introduced a new #leadership & #governance structure: https://t.co/CHNi23biLr\n\nMeet the #SteeringGroup at https://t.co/qIvWku2Dyb\n\nWe\u2019re thinking about:\n\n\u260e\ufe0f clearer communication pathways\n\ud83d\udcaa building resilience\n\ud83d\udc93 maintaining & sustaining our community\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlsPayWAAAYKLD.png"}]}, {"id": "1241142378632818693", "user": "BIDSstandard", "date": "2020-03-20T23:19:37+00:00", "text": "(6) Join the #BIDS community! You can find lots of information at https://t.co/40tEijPYro, here on twitter  and on .\n\nWe\u2019d love for you to please complete this short survey on how you want to #engage with us! https://t.co/dEHFnAwKsG\n\n#WeWantYou #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETlsXsGXkAI3cwJ.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-20T23:15:06+00:00", "text": "#OHBMx-000 \u2733 #break\n\n\u25b6 COFFEE BREAK https://t.co/EW1NOPV80B", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlqWdcXsAQVcEm.jpg"}], "ids": ["1241141243062648833"], "thread": []}, {"date": "2020-03-20T23:30:08+00:00", "text": "#OHBMx-73 \u2733 #talk\n\nOHBM Trainees @OHBM_Trainees*, K Sitek @krsitek, A Haugg @amhaugg, M Gao @Mengxia_Gao\n\n*OHBM\n\n\u25b6 Advocating for the career and personal needs of students and postdocs in the neuroimaging community", "media": [], "ids": ["1241145025888739329"], "thread": [{"id": "1241146036200116225", "user": "OHBM_Trainees", "date": "2020-03-20T23:34:09+00:00", "text": "equinoX    1/ Hello all! We\u2019re the @OHBM Student\u2013Postdoc Special Interest Group! We promote opportunities for professional, personal & career development and advocate for the needs of students and postdocs in the brain imaging community https://t.co/vzlQV4gaaa #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlvY9RXgAEM3wk.jpg"}]}, {"id": "1241146736447492096", "user": "OHBM_Trainees", "date": "2020-03-20T23:36:56+00:00", "text": "equinoX    @OHBM        2/ First, help us advocate on your behalf! What are YOUR biggest needs as an early career researcher? Comment with more details, or DM us if that\u2019s more comfortable! #OHBMx", "media": []}, {"id": "1241147375378468865", "user": "OHBM_Trainees", "date": "2020-03-20T23:39:28+00:00", "text": "equinoX    @OHBM        3/ One of our biggest focuses is supporting career development. At the annual @OHBM meeting, we host several career-related workshops on topics that are particularly relevant for students and postdocs: https://t.co/Jawe8qXM4N  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlw5bQXgAIeqm1.jpg"}]}, {"id": "1241147987843256321", "user": "OHBM_Trainees", "date": "2020-03-20T23:41:54+00:00", "text": "equinoX    @OHBM        4/ In addition, beginning in 2019, @ohbmtrainees started hosting #metoo workshops on conflict resolution, power dynamics, and handling uncomfortable situations featuring : https://t.co/CLfaRoPowF #OHBMx", "media": []}, {"id": "1241148605559377922", "user": "OHBM_Trainees", "date": "2020-03-20T23:44:21+00:00", "text": "equinoX    @OHBM         5/ We also host our annual @OHBM Mentorship Symposium with career-related keynote talks. Last year we heard from  &  about failure & success, and from  on differences between science goals & career goals. https://t.co/oxKrSg1tR2 #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlx8OgXgAARbdQ.jpg"}]}, {"id": "1241149882896920576", "user": "OHBM_Trainees", "date": "2020-03-20T23:49:26+00:00", "text": "equinoX    @OHBM            6/ Following the mentorship symposium is a lunch with mentors, giving trainees direct networking and mentoring opportunities with successful scientists. \n\nWho here has joined us for our mentor lunches?\n\nShout out some helpful mentors/advice here!\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETlzLJtXQAA1W32.jpg"}]}, {"id": "1241150431012806659", "user": "OHBM_Trainees", "date": "2020-03-20T23:51:36+00:00", "text": "equinoX    @OHBM            7/ But our mentorship efforts actually run year-round! Our online mentorship program pairs mentors with mentees\u2014not only students and postdocs, but also early stage PIs: https://t.co/OBEqDrC8sW.\n \nYou can listen to our mentees\u2019 feedback here: https://t.co/z9dw15PuiM \n#OHBMx", "media": []}, {"id": "1241151084611211264", "user": "OHBM_Trainees", "date": "2020-03-20T23:54:12+00:00", "text": "equinoX    @OHBM            8/ To further our outreach and advocacy efforts beyond the 4-day @ohbm conference, our blog tackles mentorship, career planning, and more: https://t.co/ABynczmjXN\n\nWhat topics would you like us to cover in the coming months?\n\nDM us if you have a story you\u2019d like to tell! #OHBMx", "media": []}, {"id": "1241151751069347842", "user": "OHBM_Trainees", "date": "2020-03-20T23:56:51+00:00", "text": "equinoX    @OHBM            9/ Last\u2014but certainly not least\u2014our Monday Night Social is always a lively, fun way to get to know other trainees! (and other attendees, too!) https://t.co/1Oqqio0c8y\nGames + art + dancing = inspiration :) #ohbmx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl04VUWAAU0hSs.jpg"}]}, {"id": "1241152273180438534", "user": "OHBM_Trainees", "date": "2020-03-20T23:58:56+00:00", "text": "equinoX    @OHBM            10/ Thank you for following along with us! We want to help you do your best science and live your best, most impactful lives. Follow us, tell us how we can help you, and stick up for each other out there  \ud83d\ude03https://t.co/vzlQV4gaaa #OHBMx", "media": []}, {"id": "1241432161615568903", "user": "OHBM_Trainees", "date": "2020-03-21T18:31:06+00:00", "text": "equinoX    @OHBM        Thanks for your input! A majority of responders said that career mentorship is their biggest need.\n\nAs you can see from the rest of our thread, career development\u2014both in & out of academia\u2014is a major   focus. With your feedback we'll make sure it stays a priority!", "media": []}]}, {"date": "2020-03-21T00:00:04+00:00", "text": "#OHBMx-74 \u2733 #talk\n\nElvisha Dhamala @elvisha9*, K Jamison, A Kuceyeski @AmyKuceyeski\n\n*Weill Cornell Medicine\n\n\u25b6 Hybrid Structure-Function Connectome Predicts Sex\n\n#Connectivity #Networks #Sex", "media": [], "ids": ["1241152560016302081"], "thread": [{"id": "1241152833740734464", "user": "elvisha9", "date": "2020-03-21T00:01:09+00:00", "text": "1/6 Insight into sex differences in healthy \ud83e\udde0 structure and function provides a foundation for \n1) delineating sex-specific pathology in neurological disorders displaying sex differences in clinical profiles & \n2)  guiding the development of sex-specific treatment. \n#OHBMx-74", "media": []}, {"id": "1241153641735733248", "user": "elvisha9", "date": "2020-03-21T00:04:22+00:00", "text": "2/6 Functional connectivity (FC) represents similarity of neural activation over time within \ud83e\udde0 regions.\nStructural connectivity (SC) represents white matter tracts connecting \ud83e\udde0 regions.\nSC & FC integration\u27a1\ufe0fstructurally-weighted FC (SWFC) & hybrid connectivity (HC).\n#OHBMx-74", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl2MN4XkAYERgi.jpg"}]}, {"id": "1241154390955823106", "user": "elvisha9", "date": "2020-03-21T00:07:20+00:00", "text": "3/6 Here we use SC, FC, SWFC, and HC from brain volume-matched \u2642\ufe0f-\u2640\ufe0f pairs (n=364, ) to predict sex and identify features predictive of sex using a linear support vector machine.\n #OHBMx-74", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl25RzXYAI9ENX.png"}]}, {"id": "1241155008353185792", "user": "elvisha9", "date": "2020-03-21T00:09:48+00:00", "text": "4/6 Sex classification using HC (AUC=0.98, accuracy=91%) significantly outperforms (p<0.05) FC (0.94, 87%), SC (0.96, 88%), and SWFC (0.94, 85%) in terms of AUC. \n#OHBMx-74", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl31L1XgAAHhVk.jpg"}]}, {"id": "1241155396607389701", "user": "elvisha9", "date": "2020-03-21T00:11:20+00:00", "text": "5/6 FC & SC feature importance for sex classification show striking complementarity (r=0.001, p=0.83). \n\n\u203c\ufe0f Most important connections \u203c\ufe0f:\nFunctional: to/from frontal & occipital lobes, cerebellum \nStructural: to/from parietal & temporal lobes, subcortical areas\n#OHBMx-74", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl4Ic1XYAEw5Mp.jpg"}]}, {"id": "1241155975781986306", "user": "elvisha9", "date": "2020-03-21T00:13:38+00:00", "text": "6/6 Our findings demonstrate that FC- and SC- based predictions are complementary, and suggest that integration of multimodal data, such as in HC, is crucial to understanding neurophysiological sex differences in the \ud83e\udde0.  \n#SABV\n#OHBMx-74", "media": []}]}, {"date": "2020-03-21T00:15:01+00:00", "text": "#OHBMx-75 \u2733 #talk\n\nDavide Valeriani @DavideValeriani*\n\n*Harvard Medical School\n\n\u25b6 Merging humans and machines with collaborative brain-computer interfaces\n\n#Decision", "media": [], "ids": ["1241156324056076288"], "thread": [{"id": "1241156612141625344", "user": "DavideValeriani", "date": "2020-03-21T00:16:10+00:00", "text": "1) Recognizing a person in a crowded environment is challenging for both humans and machines. How can we use neurotechnologies and machine learning to make better decisions? Well, we can make cyborgs! \ud83e\udd16 #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETl5TCqXQAAn0Cy.mp4", "content-type": "video/mp4"}]}, {"id": "1241157096445546499", "user": "DavideValeriani", "date": "2020-03-21T00:18:06+00:00", "text": "2) We showed 10 human participants pictures of crowded environments for 300 ms and asked them to decide whether a target person was present (try it!). \nThe same task was also performed by a pre-trained residual neural network (ResNet, code: https://t.co/pXKvSxfSJJ). \ud83d\udc69\ud83d\udc68\ud83d\udda5\ufe0f#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETl5rajXQAEiAnY.mp4", "content-type": "video/mp4"}]}, {"id": "1241157830989828098", "user": "DavideValeriani", "date": "2020-03-21T00:21:01+00:00", "text": "3) Collaborative brain-computer interfaces (#BCI) decoded the human decision confidence (probability of correct decision) from #EEG activity using logistic regression. The ResNet confidence was estimated as difference between target and current face encodings. #OHBMx \ud83e\udde0", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl5_PyWAAEglIW.jpg"}]}, {"id": "1241158269764349952", "user": "DavideValeriani", "date": "2020-03-21T00:22:45+00:00", "text": "4) Since confidence correlates with correctness, we used it to weigh individual decisions and obtain group decisions in different types of simulated groups, with or without the BCI and the ResNet. Performance were assessed using cross-validation. #OHBMx \ud83d\udc68\u200d\ud83d\udc69\u200d\ud83d\udc67\u200d\ud83d\udc67", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl6oM7WoAEwcyg.jpg"}]}, {"id": "1241158837144621057", "user": "DavideValeriani", "date": "2020-03-21T00:25:01+00:00", "text": "5) Combining human decisions, BCIs, and ResNet (orange line) yielded significantly more accurate decisions (up to 35%) than equally-sized human groups. Also, the average group of 5+ BCI-assisted humans and the ResNet was more accurate than the ResNet alone. #OHBMx \ud83d\udcc8", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl6_yQX0AI8n11.jpg"}]}, {"id": "1241159435952816135", "user": "DavideValeriani", "date": "2020-03-21T00:27:23+00:00", "text": "6) To sum up: machines should work with us, rather than replacing us, if we want to make the best decisions! More details on https://t.co/ORzZshMLAv.\nThanks to  for advising,  for the artwork, and you all for reading! #OHBMx \ud83d\ude47\u200d\u2642\ufe0f", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl7uRuWAAEMBpk.png"}]}]}, {"date": "2020-03-21T00:30:06+00:00", "text": "#OHBMx-76 \u2733 #talk\n\nJonathan Peelle @jpeelle*, M Jones, Z Zhu, A Luor @austinluor, A Bajracharya @aahanabajra\n\n*Washington University\n\n\u25b6 Comparing approaches for motion mitigation in task-based fMRI\n\n#Methods #Tools", "media": [], "ids": ["1241160118223470597"], "thread": [{"id": "1241160748325363714", "user": "jpeelle", "date": "2020-03-21T00:32:36+00:00", "text": "1\nMotion in fMRI hurts our stats. Including 6 canonical motion parameters as covariates (still common) is probably not optimal.\n\nIn resting state, \u201cscrubbing\u201d has worked well. Maybe this is also true for task fMRI? We investigated.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl9G1fXgAE-N_n.jpg"}]}, {"id": "1241161385867911169", "user": "jpeelle", "date": "2020-03-21T00:35:08+00:00", "text": "2\nFirst, how to identify \"bad\" scans? We used framewise displacement (FD) and chose a threshold based on group data loss. For example, an average of 2% scans lost for the group (high motion subjects would lose more, low motion subjects less).\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl9dEBWoAAiMwA.jpg"}]}, {"id": "1241161776072470530", "user": "jpeelle", "date": "2020-03-21T00:36:41+00:00", "text": "3\nFor each bad scan, we added a column to the design matrix (0s at every time point except the bad scan, which is a 1). This allows us to do \"scrubbing\" from within the GLM rather than by actually discarding data.\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl9y6HWoAEkFkM.jpg"}]}, {"id": "1241162166616707079", "user": "jpeelle", "date": "2020-03-21T00:38:14+00:00", "text": "4\nWe then compared different levels of scrubbing to 6 canonical motion parameters as covariates in several openly available datasets (Thanks !). We looked at both maximum T statistics (whole brain and task-relevant ROI) and test-retest metrics in single subs\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl-H5OX0AETb_1.jpg"}]}, {"id": "1241162598726406144", "user": "jpeelle", "date": "2020-03-21T00:39:57+00:00", "text": "5\nShowing both single dataset, and several (not all!) results. In general we found some scrubbing (i.e. 1%\u20135%) typically improved maximum t values, and test-retest was typically numerically higher than with 6 motion regressors. Possibly esp. true in high motion data.\n\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETl-iFXXkAEDHV2.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETl-jfMX0AY9k-f.jpg"}]}, {"id": "1241162984396853249", "user": "jpeelle", "date": "2020-03-21T00:41:29+00:00", "text": "6\nCurrently extending to include:\n\n-more datasets\n\n-comparisons to wavelet despiking, robust weighted least squares, and 24 motion parameter regressors\n\nbut overall our feeling is scrubbing > 6 motion parameters (usually).\n\n#OHBMx", "media": []}]}, {"date": "2020-03-21T00:45:02+00:00", "text": "#OHBMx-77 \u2733 #talk\n\nMarc Coutanche @MarcCoutanche*\n\n*University of Pittsburgh\n\n\u25b6 Applications of Informational Connectivity\n\n#Connectivity #Methods #Networks #Tools", "media": [], "ids": ["1241163876919582722"], "thread": [{"id": "1241164231787057157", "user": "MarcCoutanche", "date": "2020-03-21T00:46:27+00:00", "text": "1. MVPA can detect information in distributed patterns of fMRI activity. Functional connectivity measures shared fluctuations over time. Informational connectivity (IC) combines these (https://t.co/V9ZSx3yRU8). IC has now been applied to various Qs. Let's look at some!\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmAIEZX0AMu5GM.png"}]}, {"id": "1241164810122919937", "user": "MarcCoutanche", "date": "2020-03-21T00:48:45+00:00", "text": "2.  & Stark examined categorical representations in MTL cortex (https://t.co/h3gBSGdX9N). IC between PHC & RSC/PCC was significant for faces vs scenes, not faces vs objects - suggesting stimulus-dependent modulation of their common pattern discriminability\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmAhpPX0AEuN79.jpg"}]}, {"id": "1241165330522812420", "user": "MarcCoutanche", "date": "2020-03-21T00:50:49+00:00", "text": "3.  & Turk-Browne asked how attention promotes episodic encoding (https://t.co/P1bYwW6BTA). IC (for attentional state) between CA2/CA3/DG & RSC was greater for later remembered trials - perhaps showing how visuospatial info is transformed to LT memory\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmBSKdWoAENDlD.jpg"}]}, {"id": "1241165704423968768", "user": "MarcCoutanche", "date": "2020-03-21T00:52:18+00:00", "text": "4.  et al asked how hippocampus supports both episodic encoding & integration across experiences (https://t.co/B9eLFjO0Te). Laminar & direct MTL links had stronger IC than did indirect. Laminar connections may reflect direct flow of info between EC layers\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmBbBeWsAskRHR.jpg"}]}, {"id": "1241166050894413830", "user": "MarcCoutanche", "date": "2020-03-21T00:53:40+00:00", "text": "5. Finally, with Thompson-Schill, I investigated how learning real-world size of concepts changes early visual patterns (https://t.co/V9ZSx3yRU8). IC between EVC & angular gyrus for similarly-sized animals increased after learning, perhaps from new top-down info\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmBtJzXQA8Qb1E.jpg"}]}, {"id": "1241166441442938880", "user": "MarcCoutanche", "date": "2020-03-21T00:55:14+00:00", "text": "6. To summarize, IC can tell us a lot about how regions work together, particularly in exchanging fine-grained (MVP) information. I\u2019m looking forward to future examples! (If you prefer a video overview of method: https://t.co/zfc9QYauK9). Questions always welcome!\n#OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETmCTAwXkAUiNXk.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-21T01:00:03+00:00", "text": "#OHBMx-78 \u2733 #talk\n\nCeren Tozlu @crntozlu*, K Jamison, S Gauthier, A Kuceyeski @AmyKuceyeski\n\n*Department of Radiology-Weill Cornell Medicine\n\n\u25b6 Brain connectivity predicts MS patients\u2019 impairment level\n\n#Connectivity #Dynamics #Modeling #Networks", "media": [], "ids": ["1241167654498566144"], "thread": [{"id": "1241168283203706882", "user": "crntozlu", "date": "2020-03-21T01:02:33+00:00", "text": "1/ In Multiple Sclerosis (MS), the correlation between the impairment and disease burden measured with conventional MRI is poor. Therefore, structural and static/dynamic functional connectivity (SC,sFC,&dFC) may be used to better understand variability of impairment in MS. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmD6TiWkAE7_zb.jpg"}]}, {"id": "1241169229312856066", "user": "crntozlu", "date": "2020-03-21T01:06:18+00:00", "text": "2/ Aims:\n-to assess how well an ensemble model consisting of SC, sFC, and dFC can classify MS vs healthy controls (HC), and MS patients with clinically significant vs non-significant impairment (CSI vs CNSI)\n-to identify the relative contribution of brain\u2019s connectivity. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmEuEpXYAYKxiS.jpg"}]}, {"id": "1241169727923388421", "user": "crntozlu", "date": "2020-03-21T01:08:17+00:00", "text": "3/ Age and gender matched 76 MS patients (23 CSI and 53 CNSI) and 15 HC were included. \nHard-clustering and fuzzy meta-state analyses were used for dFC approach. \nAn ensemble model with Random Forest was used to classify \n-HC vs MS \n-MS patients with CSI vs CNSI. #OHBMx", "media": []}, {"id": "1241170011219263490", "user": "crntozlu", "date": "2020-03-21T01:09:25+00:00", "text": "4/ The ensemble model that averaged the predictions from SC and sFC better distinguished MS vs HC (AUC=0.78). For the classification of CSI vs CNSI, the SC and dFC combination performed better (AUC=0.65) compared to the combination of SC and sFC (AUC=0.60). #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmFdwnWsAAIRNp.jpg"}]}, {"id": "1241170912143200263", "user": "crntozlu", "date": "2020-03-21T01:12:59+00:00", "text": "5/ The most important connections to classify \n-MS vs HC: SC between left accumbens & left putamen and FC between right putamen & right inferior parietal\n-CSI vs CNSI: SC between left ventral DC & left cerebellum and FC between left frontal pole & right paracentral. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmFnqlXgAAPoAO.jpg"}]}, {"id": "1241171380806303744", "user": "crntozlu", "date": "2020-03-21T01:14:51+00:00", "text": "6/ dFC better classified MS with CSIvsCNSI compared to sFC. Ensemble model has potential to help clinicians accurately predict impairment in MS and identify important SC&FC underlying these impairments, thus enabling personalized prognostics and therapeutic interventions. #OHBMx", "media": []}]}, {"date": "2020-03-21T01:15:10+00:00", "text": "#OHBMx-79 \u2733 #talk\n\nMaggie Mae Mell @MaggieMaeWokeUp*, G St-Yves, T Naselaris\n\n*Medical University of SC\n\n\u25b6 Investigating unexplained variance with voxel-to-voxel models\n\n#Connectivity #Modeling #Sensory", "media": [], "ids": ["1241171457692098561"], "thread": [{"id": "1241171773246320640", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:16:25+00:00", "text": "1 Best encoding models for predicting fMRI responses in visual areas are based on Deep Neural Nets. But even DNN-based models fail to accurately predict activity to natural scenes in most voxels. Is the source of missing variance just noise or are models missing something? #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmHAlRXgAA_O5o.jpg"}]}, {"id": "1241172314286370819", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:18:34+00:00", "text": "2 Hypothesis: A portion of missing variance is due to endogenous processes. Feedback from areas higher in or outside of visual hierarchy are not accounted for in DNN stimulus-to-voxel models. We use voxel-to-voxel models (e.g. activity in V1 predicts V2) to investigate. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmHSPKWkAYCLO1.jpg"}]}, {"id": "1241172899786108928", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:20:53+00:00", "text": "3 Vox2vox > stim2vox prediction accuracy in nearly all source/target pairs. So activity in V4, while cortically \u2018distant\u2019, can more accurately predict V1 activity than stimulus features can. We infer source of unexplained variance is shared within & between visual areas. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmHvedWoAEzi9r.jpg"}]}, {"id": "1241173582601994240", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:23:36+00:00", "text": "4 Vox2vox <= stim2vox prediction accuracy when source areas in one sub\u2019s brain predict target voxel activity in a different subject. Indicates cross-subject vox2vox models are, like stim2vox models, blind to a source of variance that is common to voxels in the same brain. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmITtmXQAACQMn.jpg"}]}, {"id": "1241174177979289604", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:25:58+00:00", "text": "5 Source voxels w/large positive vox2vox model weights had receptive field (rf) locations that cluster near target voxels\u2019 rf loc. Can use weights to accurately estimate \u201cground truth\u201d rf loc of target voxels, including ones stim2vox fails on. Missing variance != all noise #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmJI4TWoAYqpoQ.jpg"}]}, {"id": "1241174942911250436", "user": "MaggieMaeWokeUp", "date": "2020-03-21T01:29:00+00:00", "text": "6 Conclusion: Best models don't explain a lot of variance. Do we see evidence this variance is endogenous & not noise?\nVariance shared across voxels.\u2705\nBut NOT shared across subs.\u2705\nRetinotopically mapped across long cortical distances.\u2705\nPreprint: https://t.co/pk2zcbFUIK\n#OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmJc3NWoAAv2Ga.jpg"}]}]}, {"date": "2020-03-21T01:30:04+00:00", "text": "#OHBMx-80 \u2733 #talk\n\nIdan Blank @IdanAsherBlank*, E Fedorenko @ev_fedorenko\n\n*UCLA\n\n\u25b6 An alternative to \u201crandom effects\u201d with higher validity, reliability, and power\n\n#Language #Memory #Methods", "media": [], "ids": ["1241175209744441349"], "thread": [{"id": "1241175643909324800", "user": "IdanAsherBlank", "date": "2020-03-21T01:31:48+00:00", "text": "1\ufe0f\u20e3How do u pool fMRI data across ppl? Most still use voxelwise pooling in common space: group-based, random-effects (RFX) analysis. But it works only if same voxel = same function across ppl, which does not hold! (https://t.co/6bvFtdeZBZ) Fig: why RFX is really bad for you #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmKplGUEAAvT29.jpg"}]}, {"id": "1241176136907780102", "user": "IdanAsherBlank", "date": "2020-03-21T01:33:45+00:00", "text": "2\ufe0f\u20e3U can use a task to find FUNCTIONAL regions in each individual brain. But to eyeball maps (\u201cblob A in brain X ~ blob B in brain Y\u201d) is unprincipled, & hard if activity is a swath (not blobs). So @ev_fedorenko devised group-constrained, subject-specific localization (GSS) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmLHOjVAAEzhlK.jpg"}]}, {"id": "1241176694850867200", "user": "IdanAsherBlank", "date": "2020-03-21T01:35:58+00:00", "text": "3\ufe0f\u20e3GSS should be better (https://t.co/jL7IKncg4B). So we pitted it against RFX! Tested 5 networks differing in laterality, domain- & modality-specificity/generality: language (n=510 ppl), Multiple Demand (392), Default Mode (386), Theory of Mind (132), face processing (102) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmLm2_UMAAkwGI.png"}]}, {"id": "1241177399825326080", "user": "IdanAsherBlank", "date": "2020-03-21T01:38:46+00:00", "text": "4\ufe0f\u20e3First we generated broad masks to constrain fROI location (top Fig). GSS = fROIs may slightly vary in location across ppl as long as within same mask. RFX = group-based fROIs (traditional). For both, effect sizes extracted w/ across run cross validation (CV; bottom Fig.) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmMNZQUMAAmKM7.jpg"}]}, {"id": "1241177870359080960", "user": "IdanAsherBlank", "date": "2020-03-21T01:40:38+00:00", "text": "5\ufe0f\u20e3GSS ALWAYS outperformed RFX, & by a huge margin! Note: even w/ n>100, RFX totally misses some regions (0 activations, see arrows) but GSS finds robust effects; in MD, classic response is \"hard>easy\", but GSS-easy > RFX-hard. So RFX massively underestimates effects #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmMqPVUYAEw5XA.png"}]}, {"id": "1241178567767027716", "user": "IdanAsherBlank", "date": "2020-03-21T01:43:25+00:00", "text": "6\ufe0f\u20e3GSS is principled, easy (toolbox: https://t.co/4akyI1ncUa) & CHEAP (for our tasks, >80% power w/ n=10). Results are valid (recall Fig1) & reliable (CV built-in). Even if you designed your expt w/o thinking GSS, you can apply it post-hoc! Cool, huh? Please stop using RFX. #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmNUs7VAAEu74W.jpg"}]}, {"id": "1241178741616726016", "user": "IdanAsherBlank", "date": "2020-03-21T01:44:06+00:00", "text": "Apologies for the broken links. Here are the correct ones:\n1. https://t.co/CaAGkmqJh9\n2. https://t.co/tH1vbXCtZu", "media": []}]}, {"date": "2020-03-21T01:45:04+00:00", "text": "#OHBMx-81 \u2733 #talk\n\nKulpreet Cheema @kulcheema*, A T Shafer @BrainHacktivist, R Bahktiari, M Moore, F Dolcos, A Singhal\n\n*Department of Neuroscience, University of Alberta\n \n\u25b6 Connectivity of prefrontal regions associated differently with emotional processing\n\n#Attention #Connectivity #Emotion", "media": [], "ids": ["1241178984899100672", "1241179074850164737"], "thread": [{"id": "1241179582964760577", "user": "kulcheema", "date": "2020-03-21T01:47:27+00:00", "text": "\ud83d\udc4b\n1. Task MRI studies show that dorsal executive (DES; hub in dlPFC) and ventral effective systems (VES; hub in vlPFC) work in opposition to each other \n\nGoal: study the brain-behaviour relationship between resting connectivity of dlPFC and vlPFC and emotional behaviours #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmOF58U4AAeLCX.jpg"}]}, {"id": "1241180093910626304", "user": "kulcheema", "date": "2020-03-21T01:49:29+00:00", "text": "2. 19 subjects completed Emotional Contagion Scale (sensitivity to other\u2019s emotions) and Emotional Regulation questionnaire + 7-minute resting-state scan \n\nSeed-to-voxel connectivity maps of bilateral dlPFC and vlPFC seed regions were correlated with the behavioural scores #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmOViVUwAAx49C.jpg"}]}, {"id": "1241180593343172609", "user": "kulcheema", "date": "2020-03-21T01:51:28+00:00", "text": "3.Results: At rest, dlPFC (Fig 1 for connectivity maps) and vlPFC (Fig 2) are positively connected to the corresponding DES and VAS areas, respectively. \n\nInterestingly, dlPFC was also negatively connected to default mode network regions (e.g., posterior cingulate cortex)  #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmO28vUYAEZ-Sp.jpg"}, {"type": "image", "url": "https://pbs.twimg.com/media/ETmO29WU4AElcUC.jpg"}]}, {"id": "1241181048840417280", "user": "kulcheema", "date": "2020-03-21T01:53:16+00:00", "text": "4. Evidence for double-dissociation in brain-behavior relationships: \n\nIncreased connectivity of dlPFC (DES) was associated with higher emotion regulation (ERQ), while increased anti-correlation of dlPFC was associated with lower emotion contagion (ECS; attached figures)", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmPWtdUYAAtnrI.jpg"}]}, {"id": "1241181637758476288", "user": "kulcheema", "date": "2020-03-21T01:55:37+00:00", "text": "5. On the other hand, the increased connectivity of the vlPFC (VES) meant increased emotion contagion, while increased anti-correlation (i.e., separation) between vlPFC and left angular gyrus resulted in decreased emotion regulation (attached figures) #OHBMx", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmPvlUUcAEYAkT.jpg"}]}, {"id": "1241182323745230848", "user": "kulcheema", "date": "2020-03-21T01:58:20+00:00", "text": "6. Patterns at rest of the key nodes of DES and VAS networks are related to how individuals interact with and handle emotional information \n\nFuture directions include studying the temporal dynamics of emotion-cognition interaction through simultaneous fMRI-EEG analyses #OHBMx", "media": [{"type": "video", "url": "https://video.twimg.com/tweet_video/ETmQvd6UMAAx53g.mp4", "content-type": "video/mp4"}]}]}, {"date": "2020-03-21T02:00:05+00:00", "text": "The #OHBMx US hub wishes to thank you for attending the conference! The US hub took precautions due to COVID-19. @dr_sheraz was in a virtual Hawaii, and @dimitrpantazis was continuously running antivirus in his tablet to prevent an infection! ", "media": [{"type": "image", "url": "https://pbs.twimg.com/media/ETmRznKXgAERjPb.jpg"}], "ids": ["1241182763782410242", "1241183504639164417"], "thread": []}, {"date": "2020-03-21T02:00:35+00:00", "text": "Special thanks belong to our keynote tweeters: @drBreaky, @bttyeo, @Lauri_Parkkonen, @fmrib_karla, @lauradata and @LucinaUddin", "media": [], "ids": ["1241182888865009668"], "thread": []}, {"date": "2020-03-21T02:01:15+00:00", "text": "This concludes the 2020 @OHBMequinoX conference. Thanks for participating, everybody. It has been an amazing twitter conference day full of interesting presentations! Don\u2019t forget to retweet your favorite ones! \n\nThe @OHBMequinoX team wishes everybody good night! #OHBMx", "media": [], "ids": ["1241183057383755778"], "thread": []}, {"date": "2020-03-23T15:58:24+00:00", "text": "Missed the first #OHBMx twitter conference? You can catch up with this amazing tool created by @anibalsolon, browse all the talks in an easier to digest format. Thanks @anibalsolon! \n\nhttps://t.co/AcorVUjHZS\nhttps://t.co/5V4Kt78Kgn\u2026", "media": [], "ids": ["1242118507829112833"], "thread": []}]}