The Past, Present, and Future of the Brain Imaging Data Structure (BIDS)

The Pas , P esen , and Fu u e o he B ain Imaging Da a S uc u e (BIDS) Pold ack, Russell A.1, Ma kiewicz, Ch is ophe J.1, Appelho , S e an2, Asha , Yoni K.3, Aue , Tibo 4,5, Baille , Syl ain6, Bansal, Shashank7, Bel achini, Leand o8, Bena , Ch is ian G.9, Be azzoli, Giacomo10,11,12,13,14, Bhogawa , Suyash15, Blai , Ross W.1, Bo ole o, Ma a10, Boud eau, Ma hieu16, B ooks, Teon L.1, Calhoun, Vince D.17, Cas elli, Filippo Ma ia18,19, Clemen , Pa icia20,21, Cohen, Alexande L.22, Cohen-Adad, Julien16, D'Amb osio, Sasha23,24, de Hollande , Gilles25, de la Iglesia-Vayá, Ma ía26, de la Vega, Alejand o27, Delo me, A naud28, De insky, O in29, D aschkow, Dejan30, Du , Eugene Paul31, DuP e, Elizabe h1, Ea l, E ic32, Es eban, Osca 33, Feingold, F anklin W.1, Flandin, Guillaume34, Galassi, An hony32, Galli o, Giuseppe35,36, Ganz, Melanie37,38, Gau, Rémi39, Gholam, James40, Ghosh, Sa aji S.41, Giacomel, Alessio42, Gillman, Ashley G.43, Gleeson, Pad aig44, G am o , Alexand e45, Guay, Samuel46, Guidali, Giacomo47, Halchenko, Ya osla O.48, Handwe ke , Daniel A.32, Ha dcas le, Nell1, He holz, Pee 49, He mes, Do a50, Honey, Ch is ophe J.51, Innis, Robe B.32, Ioanas, Ho ea-Ioan48, Jahn, And ew52, Ka akuzu, Agah16, Kea o , Da id B.53,54,55, Kia , G ego y56, Kincses, Balin 35,36, Lai d, Angela R.57, Lau, Jona han C.58, Laza i, Albe o59, Lega e a, Jon Hai z60, Li, Adam61, Li, Xiang ui62, Lo e, B adley C.63, Lu, Hanzhang64, Ma can oni, Eleono a65, Maume , Camille66, Mazzamu o, Giacomo67, Meisle , S e en L.68, Mikkelsen, Ma k69, Mu sae s, Henk70,71, Nichols, Thomas E.72, Nikolaidis, Aki73, Nilsonne, Gus a 74,75, Niso, Guioma 76, No gaa d, Ma in32,37, Okell, Thomas W.59, Oos en eld, Robe 77,78, O , Edua d79, Pa k, Pa ick J.80, Pawlik, Ma eusz81, Pe ne , Cy il R.38, Pes illi, F anco27, Pe , Jan82, Phillips, Ch is ophe83, Poline, Jean-Bap is e84, Pollonini, Luca85,86, Raamana, P adeep Reddy87, Ri e , Pe a88,89,90,91,92, Rizzo, Gaia93,94, Robbins, Kay A.95, Rockhill, Alexande P.96, Roge s, Ch is ine97, Rokem, A iel98, Ro den, Ch is99, Rou ie , Alexand e100, Sabo i -To es, Jose Manuel26, Salo, Taylo 101, Schi ne , Michael88,89,90,91,92, Smi h, Robe E.102,103, Spisak, Tamas35,104, Sp enge , Julia105, Swann, Nicole C.106, Szin e, Ma in105, Take ka , Syl ain105, Thi ion, Be and45, Thomas, Adam G.32, To abian, Sajjad107, Va oquaux, Gael108, Voy ek, B adley109, Welzel, Julius110, Wilson, Ma in111, Ya koni, Tal112, Go golewski, K zysz o J.1 1: Depa men o Psychology, S an o d Uni e si y, S an o d, CA, USA 2: Max Planck Ins i u e o Human De elopmen , Be lin, Ge many 3: Uni e si y o Colo ado Anschu z Medical Campus, Au o a, CO, USA 4: School o Psychology, Uni e si y o Su ey, Guild o d, UK 5: A i icial In elligence and In o ma ics g oup, Rosalind F anklin Ins i u e, Ha well Campus, Didco , UK 6: McConnell B ain Imaging Cen e, Mon éal Neu ological Ins i u e, McGill Uni e si y, Mon éal, Canada 7: Depa men o Bioenginee ing, Uni e si y o Cali o nia, San Diego, La Jolla, CA, USA 8: Ca di Uni e si y B ain Resea ch Imaging Cen e (CUBRIC), School o Physics and As onomy, Ca di Uni e si y, Wales, UK 9: Aix Ma seille Uni e si é, INSERM, INS, Ins Neu osci Sys , Ma seille, F ance 10: Neu ophysiology Lab, IRCCS Is i u o Cen o San Gio anni di Dio Fa ebene a elli, B escia, I aly 11: Cen e o Mind/B ain Sciences - CIMeC, Uni e si y o T en o, Ro e e o, TN, I aly 1 12: B igham and Women’s Hospi al, Bos on, MA, USA 13: Massachuse s Gene al Hospi al, Bos on, MA, USA 14: Ha a d Medical School, Bos on, MA, USA 15: Rackspace Technology, San An onio, TX, USA 16: Neu oPoly Lab, Poly echnique Mon éal, Mon éal, Quebec, Canada 17: T i-ins i u ional Cen e o T ansla ional Resea ch in Neu oimaging and Da a Science (TReNDS), Geo gia S a e, Geo gia Tech, Emo y, A lan a, GA, USA 18: Eu opean Labo a o y o Non-Linea Spec oscopy (LENS), Uni e si y o Flo ence, Ses o Fio en ino, I aly 19: Bio e ics s l, Cesena, I aly 20: Depa men o Medical Imaging, Ghen Uni e si y Hospi al, Ghen , Belgium 21: Depa men o Diagnos ic Sciences, Ghen Uni e si y, Ghen , Belgium 22: Depa men o Neu ology, Bos on Child en's Hospi al, Bos on, MA, USA 23: Dipa imen o di Scienze della Salu e dell'Uni e si à degli S udi di Milano, Milan, I aly 24: Depa men o Clinical and Expe imen al Epilepsy, Uni e si y College London, UK 25: Zu ich Cen e o Neu oeconomics, Depa men o Economics, Uni e si y o Zu ich, Zu ich, Swi ze land 26: UMIB-FISABIO, Valencia, Spain 27: The Uni e si y o Texas a Aus in, Aus in, TX, USA 28: SCCN, Uni e si y o Cali o nia, San Diego, La Jolla CA USA 29: Depa men o Neu ology, NYU Langone Medical Cen e , New Yo k, NY, USA 30: Depa men o Expe imen al Psychology, Uni e si y o Ox o d, Ox o d, UK 31: UK Demen ia Resea ch Ins i u e, Depa men o B ain Sciences, Impe ial College London, London, UK 32: In amu al Resea ch P og am, Na ional Ins i u e o Men al Heal h, Be hesda, MD, USA 33: Depa men o Radiology, Lausanne Uni e si y Hospi al and Uni e si y o Lausanne, Lausanne, Swi ze land 34: Wellcome Cen e o Human Neu oimaging, Uni e si y College London, London, England, UK 35: Cen e o T ansla ional Neu o- and Beha io al Sciences, Uni e si y Medicine Essen, Essen, Ge many 36: Depa men o Neu ology, Uni e si y Medicine Essen, Essen, Ge many 37: Depa men o Compu e Science, Uni e si y o Copenhagen, Copenhagen, Denma k 38: Neu obiology Resea ch Uni , Copenhagen Uni e si y Hospi al, Copenhagen, Denma k 39: O igamin Lab, The Neu o, McGill Uni e si y, Mon eal, Quebec, Canada 40: Ca di Uni e si y B ain Resea ch Imaging Cen e (CUBRIC), School o Psychology, Ca di Uni e si y, Wales, UK 41: Massachuse s Ins i u e o Technology, Camb idge, MA, USA 42: Depa men o Neu oimaging, Ins i u e o Psychia y, Psychology and Neu oscience, King’s College London, London, England, UK 43: The Aus alian e-Heal h Resea ch Cen e, Commonweal h Scien i ic and Indus ial Resea ch O ganisa ion, Towns ille, Queensland, Aus alia 44: Depa men o Neu oscience, Physiology and Pha macology, Uni e si y College London, London, England, UK 2 45: In ia, CEA, Uni e si é Pa is-Saclay, Palaiseau, F ance 46: Uni e si é de Mon éal, Mon éal, QC, Canada 47: Depa men o Psychology & Neu oMI - Milan Cen e o Neu oscience, Uni e si y o Milano-Bicocca, Milan, I aly 48: Cen e o Open Neu oscience, Depa men o Psychological and B ain Sciences, Da mou h College, NH, USA 49: McConnell B ain Imaging Cen e, Mon éal Neu ological Ins i u e, McGill Uni e si y, Mon éal, Quebec, Canada 50: Depa men o Physiology and Biomedical Enginee ing, Mayo Clinic, Roches e , MN, USA 51: Depa men o Psychological & B ain Sciences, Johns Hopkins Uni e si y, Bal imo e, MD, USA 52: Func ional MRI Labo a o y, Uni e si y o Michigan, Ann A bo , MI, USA 53: Change You B ain Change You Li e Founda ion, Cos a Mesa, CA, USA 54: Amen Clinics, Cos a Mesa, CA, USA 55: Depa men o Psychia y and Human Beha io , School o Medicine, Uni e si y o Cali o nia, I ine, CA, USA 56: Cen e o Da a Analy ics, Inno a ion, and Rigo , Child Mind Ins i u e, New Yo k, NY USA 57: Depa men o Physics, Flo ida In e na ional Uni e si y, Miami, FL, USA 58: Depa men o Clinical Neu ological Sciences, Wes e n Uni e si y, London, On a io, Canada 59: Wellcome Cen e o In eg a i e Neu oimaging, FMRIB, Nu ield Depa men o Clinical Neu osciences, Uni e si y o Ox o d, Ox o d, UK 60: Depa men o Radiology, B igham and Women's Hospi al, Mass Gene al B igham/Ha a d Medical School, Bos on, MA, USA 61: Columbia Uni e si y, New Yo k, NY, USA 62: Cen e o Cogni i e and Beha io al B ain Imaging, The Ohio S a e Uni e si y, Columbus, OH, USA 63: Uni e si y College London, London, UK 64: Johns Hopkins Uni e si y School o Medicine, Bal imo e, MD, USA 65: School o Psychology and Neu oscience and Cen e o Cogni i e Neu oimaging, Uni e si y o Glasgow, Glasgow 66: In ia, Uni Rennes, CNRS, Inse m, IRISA UMR 6074, Empenn ERL U 1228, Rennes, F ance 67: Na ional Resea ch Council - Na ional Ins i u e o Op ics (CNR-INO), Flo ence, I aly 68: P og am in Speech and Hea ing Bioscience and Technology, Ha a d Uni e si y, Camb idge, MA, USA 69: Depa men o Radiology, Weill Co nell Medicine, New Yo k, NY, USA 70: Radiology and Nuclea Medicine, V ije Uni e si ei Ams e dam, Ams e dam UMC loca ion VUmc, Ams e dam, The Ne he lands 71: Ams e dam Neu oscience, B ain Imaging, Ams e dam, The Ne he lands 72: Big Da a Ins i u e, Li Ka Shing Cen e o Heal h In o ma ion and Disco e y, Nu ield Depa men o Popula ion Heal h, Uni e si y o Ox o d, Ox o d, UK 73: Cen e o he De eloping B ain, Child Mind Ins i u e, New Yo k, NY, USA 74: Depa men o Clinical Neu oscience, Ka olinska Ins i u e , S ockholm, Sweden 75: Swedish Na ional Da a Se ice, Go henbu g Uni e si y, Go henbu g, Sweden 3 76: Ins i u o Cajal, CSIC, Mad id, Spain 77: Donde s Ins i u e o B ain, Cogni ion and Beha iou , Radboud Uni e si y Nijmegen, Nijmegen, The Ne he lands 78: Na MEG, Ka olinska Ins i u e , S ockholm, Sweden 79: Hein ich Heine Uni e si y, Depa men o Biological Psychology o Decision Making, Düsseldo , Ge many 80: Wes e n Uni e si y, London, On a io, Canada 81: Pa is-Lod on-Uni e si y o Salzbu g, Depa men o Psychology, Cen e o Cogni i e Neu oscience, Salzbu g, Aus ia 82: Helmhol z-Zen um D esden-Rossendo , Ins i u e o Radiopha maceu ical Cance Resea ch, D esden, Ge many 83: GIGA CRC in i o imaging, Liege Uni e si y, Liege, Belgium 84: Neu o Da a Science ORIGAMI Labo a o y, McConnell B ain Imaging Cen e, Facul y o Medicine, McGill Uni e si y, Mon éal, Canada 85: Depa men o Enginee ing Technology, Uni e si y o Hous on, Hous on, TX 86: Basque Cen e on Cogni ion, B ain and Language, Donos ia-San Sebas ián, Spain 87: Uni e si y o Pi sbu gh, Pi sbu gh, PA, USA 88: Be lin Ins i u e o Heal h a Cha i é, Uni e si ä smedizin Be lin, Cha i épla z 1, Be lin 10117, Ge many 89: Depa men o Neu ology wi h Expe imen al Neu ology, Cha i é, Uni e si ä smedizin Be lin, Co po a e membe o F eie Uni e si ä Be lin and Humbold Uni e si ä zu Be lin, Cha i épla z 1, Be lin 10117, Ge many 90: Be ns ein Focus S a e Dependencies o Lea ning and Be ns ein Cen e o Compu a ional Neu oscience, Be lin, Ge many 91: Eins ein Cen e o Neu oscience Be lin, Cha i épla z 1, Be lin 10117, Ge many 92: Eins ein Cen e Digi al Fu u e, Wilhelms aße 67, Be lin 10117, Ge many 93: In ic o, London, UK 94: Di ision o B ain Sciences, Impe ial College London, London, UK 95: Depa men o Compu e Science, Uni e si y o Texas a San An onio, San An onio, TX, USA 96: Depa men o Neu osu ge y, O egon Heal h & Science Uni e si y, Po land, OR, USA 97: McGill Cen e o In eg a i e Neu oscience (MCIN), Mon éal Neu ological Ins i u e, McGill Uni e si y, Mon éal, QC, Canada 98: Uni e si y o Washing on, Depa men o Psychology and eScience Ins i u e, Sea le, WA, USA 99: Uni e si y o Sou h Ca olina, Depa men o Psychology, Columbia, SC, USA 100: In ia, Ins i u du Ce eau - Pa is B ain Ins i u e, Pa is, F ance 101: Li espan In o ma ics and Neu oimaging Cen e (PennLINC), Depa men o Psychia y, Pe elman School o Medicine, Uni e si y o Pennsyl ania, Philadelphia, PA, USA 102: The Flo ey Ins i u e o Neu oscience and Men al Heal h, Heidelbe g, Vic o ia, Aus alia 103: The Flo ey Depa men o Neu oscience and Men al Hea h, The Uni e si y o Melbou ne, Pa k ille, Vic o ia, Aus alia 104: Ins i u e o Diagnos ic and In e en ional Radiology and Neu o adiology, Uni e si y Medicine Essen, Essen, Ge many 4 105: Ins i u de Neu osciences de la Timone (INT), UMR7289, CNRS, Aix-Ma seille Uni e si é, F ance 106: Uni e si y o O egon, Depa men o Human Physiology, Eugene, OR, USA 107: Uni e si y o Cali o nia, I ine, CA, USA 108: In ia, Uni e si é Pa is Saclay, Saclay, F ance 109: Depa men o Cogni i e Science, Halıcıoğlu Da a Science Ins i u e, and Neu osciences G adua e P og am, Uni e si y o Cali o nia, San Diego, La Jolla, CA, USA 110: Kiel Uni e si y, Kiel, Ge many 111: Uni e si y o Bi mingham, Cen e o Human B ain Heal h and School o Psychology, Bi mingham, UK 112: Google X. Moun ain View, CA USA Abs ac The B ain Imaging Da a S uc u e (BIDS) is a communi y-d i en s anda d o he o ganiza ion o da a and me ada a om a g owing ange o neu oscience modali ies. This pape is mean as a his o y o how he s anda d has de eloped and g own o e ime. We ou line he p inciples behind he p ojec , he mechanisms by which i has been ex ended, and some o he challenges being add essed as i e ol es. We also discuss he lessons lea ned h ough he p ojec , wi h he aim o enabling esea che s in o he domains o lea n om he success o BIDS. Main ex The sha ing o scien i ic esea ch da a is bene icial in nume ous ways. Fo emos , i maximizes he po en ial knowledge o be de i ed om he da a, hus maximizing he bene i o he s akeholde s who und he esea ch and he con ibu ions o esea ch pa icipan s. I also p o ides he means o esea che s o a emp o ep oduce he wo k o o he s in hei ield, which is an essen ial componen o science. Fu he , i le els he scien i ic playing ield by p o iding da a o esea che s om unde - esou ced en i onmen s o hose wi hou da a acquisi ion capabili ies, which hey can use o de elop no el analysis me hods o es new scien i ic hypo heses. Gi en he ise o machine lea ning me hods in science, ano he unsung bene i o da a sha ing is ha i p o ides la ge and mo e di e se aining da ase s, which can inc ease obus ness and dec ease o e i ing and bias. Fo all o hese easons, he sha ing o da a has become inc easingly common ac oss science, as ha e demands om unding agencies and publishe s ha da a be sha ed. Wi hin he ield o neu oimaging, da a sha ing e o s s a ed a ound 2000 wi h he MRI Da a Cen e (Van Ho n e al., 2013). Ten yea s la e i began o lou ish wi h he ad en o he In e na ional Neu oimaging Da a-sha ing Ini ia i e/Func ional Connec omes P ojec (INDI/FCP) (Mennes e al., 2013). Subsequen p ojec s ha e ocused on p ospec i e sha ing o da a, including he Human Connec ome P ojec (Van Essen e al., 2013) and he Adolescen B ain Cogni i e De elopmen (ABCD) S udy (Casey e al., 2018), which ha e had a majo impac on he ield by p o iding la ge quan i ies o neu oimaging da a o esea che s. 5 The sha ing o da a is a wo hy goal, bu only i he da a a e sha ed in a way ha makes hem FAIR (Findable, Accessible, In e ope able, and Reusable) (Wilkinson e al., 2016). One majo con ibu o o FAIRness is he use o s anda d ile o ma s, which allow esea che s o euse da a ac oss mul iple so wa e pla o ms. The ield o neu oimaging esea ch using MRI has bene i ed om he longs anding con e gence o he ield on a s anda d ile o ma o imaging da a, he Neu oimaging In o ma ics Technology Ini ia i e (NI TI) o ma (Cox, R. W., Ashbu ne , J., B eman, H., & Fissell, K., 2004), which is used by nea ly all majo MRI analysis so wa e packages. Howe e , beyond ile o ma s, he da a and associa ed me ada a mus be o ganized in such a way ha da a ecipien s can quickly and accu a ely unde s and he con en s o he da a. The bene i s o a clea , o mal o ganiza ion scheme o da a include minimizing he bu den o cu a ion o esea che s and o da a sha ing eposi o ies, educing he likelihood o e o s due o misunde s anding o misin e p e a ion o he da a, enabling he de elopmen o analysis ools ha can au oma ically u ilize he s uc u e o he da a o analyze hem app op ia ely and wi h minimal use inpu , and a o ding he abili y o au oma ically alida e he da a o de e mine whe he hey mee he s anda d (Go golewski e al., 2016). In his pape we ou line he his o y, cu en s a us, and u u e di ec ions o he B ain Imaging Da a S uc u e (BIDS), which has become a widely accep ed communi y-d i en da a s anda d wi hin he neu oscience communi y. The goal o his exposi ion is o p o ide a w i en his o y o he e en s ha ga e ise o BIDS and ou line he e en s leading o i s es ablishmen , which we hope will be o in e es o esea che s in he neu oimaging ield as well as o hose in o he ields wo king o es ablish success ul new da a s anda ds. Figu e 1. A ep oduc ion o Go golewski e al., 2016, Figu e 1, showing an example mapping om DICOM o BIDS. BIDS is a communi y-d i en s anda d o o ganizing, naming, and anno a ing neu oimaging da a ha places a hea y emphasis on human- and machine- eadabili y. Since i s ini ial publica ion, BIDS has expanded om s uc u al, unc ional 6 and di usion MRI o inco po a e o he MR me hodologies such as a e ial spin labeling and o he eco ding echnologies such as elec ophysiology. The bi h o BIDS The bi h o BIDS can be aced back ul ima ely o a social media pos by Russ Pold ack on Oc obe 17, 20141. The pos e e ed o a alk ha Ch is Go golewski had gi en a he weekly S an o d cogni i e/neu oscience semina (“F isem”). A eply o he pos by S ua Buck, hen a p og am o ice a he Lau a and John A nold Founda ion, led o a discussion ha ul ima ely esul ed in a subs an ial g an om he Founda ion o he S an o d g oup, wi h he aim o de eloping a new da a sha ing pla o m ha would supe cede he Open MRI da a sha ing pla o m (Pold ack e al., 2013) ha he g oup had p e iously un. This new pla o m would ul ima ely become he OpenNeu o a chi e (Ma kiewicz e al., 2021) and he suppo om he A nold Founda ion would help s a he wo k on BIDS. One o he majo challenges o he Open MRI p ojec had been da a cu a ion. The p ojec had de eloped an in-house da a o ganiza ion scheme (see Figu e 1 in Pold ack e al., 2013), which e lec ed common p ac ice in many labs bu was buil a ound a speci ic wo k low o ask MRI analysis based on he FSL so wa e package (Jenkinson e al., 2012). S anda diza ion o ile layou and s udy design me ada a gained ac ion in o he p ojec s in e es ed in o malizing he loading o da a; e.g., PyMVPA (Hanke e al., 2009) e sion 2.6.1 (No embe 2014), included “Di ec suppo o loading da a and design om open m i.o g-s yle da ase s”2. Since he scheme was no o mally desc ibed, esea che s wishing o deposi da a could no easily ans o m hei da a o mee i . Ins ead, da a deposi o s would send hei da a o he Open MRI eam and a cu a o wi hin he eam would wo k wi h he deposi o o ans o m he da a o mee he in o mal s anda d. The e was also no way o easily de e mine whe he his ans o ma ion was co ec , o he han unning i h ough he au oma ed wo k low and seeing whe he he wo k low an success ully. This esul ed in signi ican pe sonnel cos s and g ea ly limi ed he amoun o da a ha could be inges ed o he a chi e. This s anda d was also limi ed o a e y speci ic ype o MRI da a and analysis wo k low, and hus was no necessa ily use ul o a b oad g oup o esea che s. When Ch is Go golewski and Russ Pold ack began discussing he de elopmen o a new a chi e, hey ecognized ha i was essen ial o subs an ially shi he bu den o cu a ion om he a chi e o he da a owne s, in o de o make he a chi e inancially sus ainable in he long un. I became clea ha his would equi e a de ailed and gene al scheme o he o ganiza ion o he in ended da a ypes, and ha his scheme should suppo au oma ed alida ion so ha use s can upload da a and sha e hem immedia ely wi hou he need o manual cu a ion. As i happened, he S an o d eam was al eady pa icipa ing in he Neu oimaging Da a Sha ing Task Fo ce (o NIDASH), a collec i e e o unde he umb ella o he In e na ional Neu oin o ma ics Coo dina ing Facili y (INCF). Amongs o he ini ia i es, his g oup was de eloping he Neu oimaging Da a Model (NIDM) p ojec , an in o ma ics amewo k o he o mal desc ip ion o 2h ps://gi hub.com/PyMVPA/PyMVPA/pull/240 1h ps:// wi e .com/ usspold ack/s a us/523263185764491264; sc eensho a ailable a h ps://os .io/ha8gx 7 neu oimaging expe imen s and da ase s (Maume e al., 2016). As pa o a se ies o mee ings o ganized by INCF o u he he p og ess o NIDM and o he da a sha ing e o s, a mee ing had been planned o Janua y 2015 a S an o d in o de o discuss he de elopmen o a NIDM model o MRI expe imen s, based on he Open MRI use case. This was he mee ing a which BIDS was i s en isioned. Figu e 2. A g aphical imeline o he his o ical de elopmen o he BIDS p ojec , including impo an publica ions, mee ings, and o he de elopmen s. A imeline o he de elopmen o BIDS is p esen ed in Figu e 2. The S an o d mee ing was held Janua y 27 h-30 h, 2015, wi h signi ican suppo om he INCF3. The in-pe son a endees we e Ma hew Ab ams, Michel Dumon ie , Guillaume Flandin, Ch is Go golewski, Ka l Helme , Da id Kea o , Camille Maume , Nolan Nichols, Russ Pold ack, Jean-Bap is e Poline, A iel Rokem, and Vanessa Socha ; o he in i ees a ending emo ely included Sa aji Ghosh, Ya osla Halchenko, Michael Hanke, Da id Kennedy, Angie Lai d, Tom Nichols, and Jessica Tu ne . The mee ing s a ed wi h p esen a ions on a numbe o ongoing ele an p ojec s and hei ela ion o he Open MRI use case. The i s explici men ion o BIDS came on Day 2, when one o he subg oups was labeled as “Subg oup1: OBIDS (Open B ain Imaging Da a S uc u e) o ma p oposal (de i ed om he Open MRI o ma )”. A pho o o a whi eboa d d awing (Figu e 3) shows he in ended sepa a ion o s anda ds, wi h a di ec o y-based o ma (in ended o mos use s) and a Resou ce Desc ip ion F amewo k (RDF) based o ma (in ended o compu a ionally ad anced use s); he o me is wha would become BIDS. 3The o iginal agenda and no es om he mee ing a e a ailable a h ps://os .io/kma h/ 8 Following he mee ing in Janua y 2015, he e was a subs an ial e o o de elop a se o examples ha would be ci cula ed wi h he ini ial d a o he speci ica ion, based p ima ily upon da ase s om he Open MRI da abase. An addi ional mee ing was held a he OHBM con e ence in Honolulu in June, 2015 and consequen ly a he INCF 2015 cong ess in Cai ns, Aus alia, whe e bo h OBIDS and NIDM g oups wo ked oge he as pa o NIDASH o c oss- e ilize bo h s anda ds wi h basic me ada a o desc ibe MRI expe imen s and da a. The BIDS speci ica ion d a and 22 example da ase s we e dissemina ed o he communi y o commen s on Sep embe 21s 2015, along wi h an ea ly e sion o he Ja aSc ip -based bids- alida o 4, de eloped by Squishymedia. Squishymedia was a con ac o also esponsible o p ima y de elopmen o he OpenNeu o a chi e; i la e ended ope a ions in 2021, a e which Nell Ha dcas le (lead de elope on he OpenNeu o p ojec ) joined he S an o d eam. A NIDASH ask o ce mee ing was held in Chicago in Oc obe 2015 in ad ance o he Socie y o Neu oscience (SFN) Annual Mee ing, and a BIDS lea le 5was dis ibu ed a he SFN mee ing o p omo e he s anda d o a wide audience. Figu e 3: A snapsho o he whi eboa d a he ini ial BIDS mee ing (Janua y 27-30, 2015), ou lining he in ended sepa a ion o a di ec o y-based o ma (which would become BIDS) and a o mal RDF-based desc ip ion (which would become NIDM-Expe imen ). 5A copy a ailable om h ps://web.a chi e.o g/web/20230519223151/h ps://neu o.debian.ne /_ iles/b ochu e_bids.pd and also lis ed on h ps://cen e o openneu oscience.o g/engage. 4h ps://gi hub.com/bids-s anda d/bids- alida o / 9 In he wake o he OHBM mee ing, Tom Nichols p oposed he idea o a BIDS S ee ing G oup, which would be elec ed based on a o e by he BIDS communi y. In pa allel, a plan o communi y go e nance was de eloped by membe s o he communi y (see h ps://bids.neu oimaging.io/go e nance). The de elopmen o his p ocess was led subs an ially by F anklin Feingold om he S an o d eam, who se ed as he p incipal p ojec manage o BIDS om 2018 h ough 2022. The i s elec ion was held in Oc obe 2019. The winning sla e was chai ed by Guioma Niso including Melanie Ganz, Robe Oos en eld, Russ Pold ack, and Ki s ie Whi ake , a eam ha ep esen ed he di e si y o neu oimaging modali ies al eady hen in BIDS, including MRI, PET, MEG, and EEG. In 2021 an elec ion was held o eplace Ki s ie Whi ake , which esul ed in he elec ion o A iel Rokem. In 2022, an elec ion was held o eplace Melanie Ganz and Russ Pold ack, which esul ed in he elec ion o Ya osla Halchenko and Cy il Pe ne . In 2023, an elec ion was held o eplace Guioma Niso and Robe Oos en eld, which esul ed in he elec ion o Do a He mes and Camille Maume . The con inued success ul ope a ion o he communi y in he absence o i s ounde demons a es he s eng h o he communi y go e nance model. The go e nance p ocess also o malized he ole o he BIDS main aine s. This was spu ed by he mo e o he p ima y BIDS speci ica ion documen om a collabo a i ely edi ed documen o a websi e gene a ed om a e sion-con olled Gi Hub eposi o y, which inc eased he echnical ba ie s o con ibu ion. The ini ial main aine s’ g oup consis ed o S e an Appelho , F anklin Feingold, Ross Blai , and Ch is Ma kiewicz, who ook esponsibili y o managing he eposi o y and alida o and acili a ing con ibu ions om use s less amilia wi h wo king in Gi Hub. The wo k o he main aine s g oup includes social in as uc u e, such as he BIDS websi e and social media, as well as unning he s ee ing g oup elec ions. The main aine s g oup emains a sel -selec ed se o con ibu o s ha collec i ely make in as uc u e decisions a ound he speci ica ion, alida o , and example da ase s, acili a e addi ions o he s anda d, and ad ise he s ee ing g oup. Taylo Salo and Remi Gau joined he g oup in 2020, An hony Galassi and E ic Ea l in 2021, and Ch is ine Roge s, Nell Ha dcas le and Kim Ray in 2023. One key s eng h o BIDS is he enginee ing acumen o he scien is s in ol ed, which is e lec ed in he way ha he s anda d is now ende ed o public consump ion. Ini ially he published s anda d o he web si e and a machine- eadable JSON schema used by he Ja aSc ip alida o we e kep in sync manually. The de elopmen o a gene ic, decla a i e schema o desc ibe he s anda d (desc ibed u he below) allowed he BIDS main aine s and o he con ibu o s o de elop ools ha gene a e some c ucial speci ica ion ex and ables di ec ly om he e y same schema as he alida o and he e o e a oid con lic s be ween he published s anda d and he alida o . The uni ied schema also allowed downs eam ools, such as HeuDiCon (RRID:SCR_017427), o a oid ha dcoding he s anda d, hus making hem also mo e obus o changes o he BIDS s anda d. The p esen s a e o BIDS 16 A p esen , BIDS is a highly success ul example o a communi y-d i en s anda d o da a o ganiza ion; o ou knowledge, i is one o he only g ass- oo s s anda ds o ha e gained such b oad accep ance wi hin i s ield. This communi y suppo was ecognized by he INCF when hey endo sed BIDS as a bes p ac ice s anda d in 2018, and e-endo sed i in 2021. No only has he adop ion o BIDS con inued o g ow, bu BIDS i sel has g own o e lec de elopmen s in he ield. A massi e amoun o da a is now sha ed in he BIDS o ma . The OpenNeu o a chi e (as o June 2023) sha es da a o mo e han 34,000 indi iduals om mo e han 850 BIDS da ase s; Figu e 3 shows he consis en inc ease in he size o his da abase o e ime. The ABCD-BIDS Communi y Collec ion (Feczko e al., 2021) sha es a BIDS e sion o he ABCD da ase ha includes longi udinal da a om 11,877 child en. Ano he lens in o BIDS usage comes om he s a is ics collec ed by he MRIQC BIDS app, which s o es eleme y in o ma ion abou each MRI un (unless he use op s ou ) ha is hen sha ed ia an open web API (Es eban, Blai , e al., 2019). Figu e 4 shows he cumula i e numbe o unique images (dis inguished based on a checksum o each image) submi ed o he MRIQC web API be ween 2018 and June 2023, which demons a es a sus ained and consis en g ow h in he numbe o da ase s con e ed o BIDS. Gi en ha hese da ase s ep esen only a subse o he o al numbe o BIDS da ase s, we would es ima e ha he e a e likely o be well o e 100,000 da ase s wi h millions o images ha ha e been con e ed o BIDS a his poin . These esul s highligh he ac ha BIDS is being used by a signi ican numbe o esea che s in he communi y. Figu e 4. G owing usage o BIDS o e ime. Le : G ow h o he OpenNeu o da abase since i s incep ion in 2017, adap ed om (Ma kiewicz e al., 2021). Righ : Cumula i e numbe o unique T1-weigh ed ana omical ( 1w) and BOLD images om BIDS da ase s submi ed o he MRIQC web API (Es eban, Blai , e al., 2019) om 2018 o June 2023. Sou ce da a and code o gene a e igu es a ailable a h ps://os .io/x7 h8/. BIDS has also enabled a numbe o impo an da a in eg a ions ha g ow communi y adop ion ia he so wa e ecosys em. Fo example, he cloud pla o ms b ainli e.io (Hayashi e al., 2023) and nema .o g (Delo me e al., 2022) ha e u ilized BIDS o p o ide eady- o-use da a se ices. 17 The BIDS s anda d p o ides a common da a o ma o inges ion and exchange; combined wi h Da aLad (Halchenko e al., 2021) as a ans e mechanism, OpenNeu o da ase s a e made seamlessly a ailable o b ainli e.io and NEMAR use s. Fu he , he BIDS Apps speci ica ion (Go golewski e al., 2017) enabled BIDS-awa e pla o ms such as b ainli e.io o make hi d-pa y BIDS Apps a ailable o a la ge communi y o use s. O he open-sou ce pla o ms ha in eg a e BIDS suppo , such as LORIS (Das e al., 2011) and CBRAIN (She i e al., 2014), also con ibu e open ools and wo k lows o onboa d new use g oups o BIDS (Roge s e al., 2022) as a basis o esea ch collabo a ion. To ake one inal example, ano he BRAIN Ini ia i e® da a a chi e, DANDI (RRID:SCR_017571), has e ec i ely employed BIDS in ha mony wi h s anda ds om o he sub ields, such as Neu oda a Wi hou Bo de s (Tee e s e al., 2015) and OME-Za (Moo e e al., 2023), o acili a e in eg a ion ac oss di e se domains o neu oscience da a. One majo achie emen o he BIDS communi y in ecen yea s has been he de elopmen o a machine- eadable schema o ep esen he s anda d. The ea ly eleases o BIDS consis ed o a speci ica ion w i en in English and a alida o w i en in Ja aSc ip , which aspi ed o ha e a alida ion p ocedu e o each English ule. As he s anda d g ew o include mo e da a ypes, he numbe o people wi h he expe ise needed o adap and main ain hese pa allel ep esen a ions became anishingly small. A he same ime, each ool implemen ing BIDS ecommenda ions esul ed in ye ano he ep esen a ion o he s anda d ha equi ed upda ing. These di icul ies led o an e o o c ea e a decla a i e (i.e., non-p ocedu al) schema ha is main ained as pa o he speci ica ion documen (Figu e 5). This has ou impo an bene i s. Fi s , i allows much easie inclusion o new elemen s o he s anda d (such as new BEPs). Second, i enables he consis en implemen a ion o he alida o ac oss mul iple languages. Thi d, i makes i possible o he alida o and any downs eam ool using he schema o p o ide handling o he BIDS da ase speci ic o i s BIDS e sion. Fou h, he inclusion o he schema wi h he speci ica ion encou ages con ibu o s who wish o p opose a new ule o conside he di icul y o exp essing ha new ule wi hin he schema. The schema-based alida o has been implemen ed and is cu en ly being es ed, and is expec ed o eplace he o iginal alida o in la e 2023. I has al eady been adop ed by he DANDI and OpenNeu o da a a chi es, bo h o which con ibu ed o he e o s o de elop he schema and schema-based alida o . 18 Figu e 5. O e iew o BIDS Schema usage. In his example, BEP 030 (NIRS) in oduces a ile naming ule o he schema as pa o he BEP p ocess. The schema ule is used o ende a ile naming empla e in he speci ica ion. The BIDS Valida o uses he ule o iden i y alid NIRS da a iles while ejec ing imp ope ly named iles. Finally, hi d-pa y ools, such as a que y lib a y, may inges he upda ed schema o au oma ically gain access o new ea u es o BIDS. The u u e o BIDS As BIDS nea s i s en h anni e sa y, i s success has also led o inc easing ecogni ion o he limi a ions o he exis ing amewo k. This has in u n d i en a g owing discussion ega ding he need o a new majo e sion o BIDS (“BIDS 2.0”) ha would in oduce changes ha a e incompa ible wi h he exis ing BIDS amewo k. A dedica ed eposi o y (h ps://gi hub.com/bids-s anda d/bids-2-de el/issues) is collec ing issues, a subse o which will be chosen o BIDS 2.0. I is likely ha he discussion o a new e sion will con inue o e he nex ew yea s, gi en he signi ican di icul y ha b eaking changes would impose on ool de elope s. As an example, one con en ious issue ha would equi e esolu ion in any successo speci ica ion is he “inhe i ance p inciple”. This allows me ada a o be speci ied a a highe le el in he da a hie a chy, and be “inhe i ed” by mul iple da a iles o which i is applicable. On one hand, his in oduces a deg ee o complexi y o use comp ehension and so wa e in e ac ion wi h he BIDS speci ica ion and BIDS da ase s; on he o he , i collapses edundan in o ma ion in a manne ha is ai h ul o he hie a chical na u e o he me ada a a hand, which becomes inc easingly pe inen o da ase s o inc easing complexi y. Many such si ua ions exis whe e iden i ica ion o an issue o con en ion would no ha e been possible wi hou hands-on expe ience wi hin he BIDS ecosys em. Ano he majo challenge o he BIDS communi y is o decide whe e he s anda d should end. BIDS has al eady ex ended beyond wha is classically conside ed as “neu oimaging da a”, h ough he Mic oscopy ex ension (BEP031) and o he ex ensions including gene ic in o ma ion 19 (BEP018), eye acking (BEP020), and mo emen da a (BEP029). The success o BIDS has ga ne ed in e es om many esea che s in de eloping ela ed s anda ds unde he BIDS umb ella, bu con inued expansion o he scope o he s anda d also h ea ens o inc ease i s complexi y o a deg ee ha i becomes di icul o change. The ques ion o whe e he s anda d ends is pa icula ly ele an o con inued s anda diza ion o BIDS De i a i es and hei in e ac ion wi h so wa e ools. 1. Hie a chical complexi y: The ilesys em s uc u e o he BIDS s anda d was designed o mee he equi emen s o BIDS Raw da a, wi h da a iles a anged in an immu able hie a chy ac oss da ase s, hen subjec , (op ionally) session, and inally imaging modali y. While he e may be emendous lexibili y in ile naming wi hin his s uc u e, he hie a chy i sel is in lexible. This did no pose an issue o BEP003 Common De i a i es, as each p oposed de i a i e was a s andalone piece o da a de i ed om aw da a om a single modali y. Howe e he e a e u u e p ospec s whe e his will no longe be he case, such as de i a i es ha a e he esul o explici ly mul i-modal analysis, o da a hie a chies mo e complex han ha a o emen ioned (such as he esul o a model i ha is sp ead ac oss mul iple da a iles), which canno be ai h ully ep esen ed in a BIDS s uc u e wi h inhe i ance unde he cu en s anda d. 2. De i a i es as inpu s: F om hei incep ion, BIDS Apps we e designed on he p emise o aking as inpu a BIDS Raw da ase , and p oducing as ou pu a inal se o compu ed de i a i e da a. The in e es in BIDS De i a i es has been p incipally in acili a ion o he sha ing and unambiguous in e p e a ion o such da a. In pa allel, a numbe o BIDS Apps ha e implemen ed he abili y o iden i y and u ilize he de i a i es al eady compu ed by some o he applica ion a he han duplica ing hose equisi e calcula ions in e nally. Combining hese concep s p esen s an oppo uni y o he cons uc ion o la ge, complex, mul i-modal p ocessing pipelines inco po a ing dispa a e so wa es. Jus as many exis ing BIDS Apps a e based on cons uc ing and execu ing a di ec ed acyclic g aph o unde lying indi idual commands om di e en neu oimaging so wa e packages (Go golewski e al., 2017), la ge analysis pipelines could be based on cons uc ing and execu ing a di ec ed acyclic g aph o unde lying BIDS Apps, wi h BIDS De i a i es se ing as he means o ansla ion be ween hose Apps. 3. Exis ing oolchains: A la ge p opo ion o neu oimaging analyses a e pe o med using one o a small se o exis ing so wa e packages. Many o hese ha e long-s anding ile layou schemes and ools designed o ope a e on hei own speci ic se s o de i a i es. Con o mance o such packages o a comple ely new da a s uc u e may he e o e be oo high an expec a ion. The e a e wo key ways in which his p oblem should be conside ed. Fi s ly, any de i a i es ex ension p oposal should ideally ha e accompanying i a so wa e ool o pe o ming bidi ec ional con e sions be ween he ou pu s o one o mo e majo oolchains and he p oposed speci ica ion; his would acili a e gene a ion o con o ming da a by use s and BIDS App de elope s alike, and subsequen manipula ion / isualiza ion o sha ed da a using he o igina ing oolchain. 20 Secondly, an al e na i e app oach would be o desc ibe in he speci ica ion a way o encapsula e hose esul s in he na i e oolchain o ma and me ely anno a e hem wi h ag eed common e ms, an app oach wi h p eceden s in BEPs 015 (De i a i e mapping iles) and 035 (Mega-analysis using non-complian de i a i es), and which may do e ail wi h BEP 028 (P o enance). This would pe mi pos hoc anno a ion o de i a i es ha p eda e BIDS De i a i es and/o a e gene a ed by ools ou side o he BIDS ecosys em; and while i may sac i ice ul ima e ilesys em con en ion con o mi y, i would no p eclude he ongoing pu sui o such. Ano he di ec ion o u u e e o s is u he in eg a ion wi h o he ela ed s anda ds. Whe eas he NI TI ile o ma has become s anda d wi hin he MRI esea ch communi y, DICOM has g own in o he indus y s anda d o a wide ange o imaging modali ies (such as physiology, e c.), while add essing he many sho comings ha had o iginally u ned he neu oimaging esea ch communi y owa ds simple o ma s. As a esul , many s anda diza ion e o s ha e been duplica ed. As DICOM is he indus y s anda d and mo e da a will be a i ing in DICOMs, coo dina ion wi h de elopmen s in DICOM could help o ensu e mo e apid adop ion o new imaging sequences and e en modali ies in o he BIDS s anda d, which p o ides an umb ella o ganiza ion a he s udy le el. Con inued wo k wi h ela ed s anda ds such as NWB and OME-Za will also be impo an o a oid unnecessa y duplica ion o e o o con lic ing ecommenda ions o da ase s and a chi es on he in e ace o neu oimaging and elec ophysiology/mic oscopy. The unding o con inued BIDS de elopmen and main enance emains a challenge. The BRAIN Ini ia i e has unded many o he de elopmen s o BIDS, ei he di ec ly ( o BIDS de elopmen p ojec s) o indi ec ly ( h ough unding o da a a chi es ha ha e elied upon and con ibu ed o BIDS). Much o he wo k o de elop and main ain he s anda d is pe o med by he BIDS Main aine s; many o he main aine s a e cu en ly suppo ed by ela ed g an s, which endange s he p ojec gi en ha g an s usually ha e ime windows o h ee o i e yea s. The es ablishmen o a ounda ion o suppo BIDS could be a use ul u u e de elopmen . Lessons Lea ned Gi en he demons able success o BIDS, i is use ul o ask: Wha lessons can be lea ned ha migh be use ul o o he s anda ds p ojec s? The i s impo an poin o acknowledge is ha he success o any pa icula p ojec de i es in pa om pu e luck, so one should no o e i oo hea ily o he ollowing. None heless, we belie e ha he e a e se e al po en ially impo an lessons o be lea ned. Main ac o s in he success o BIDS Absence o exis ing solu ions. Many s anda diza ion p ojec s commence on he p emise ha he e a e mul iple exis ing compe ing s anda ds in ha domain, each wi h hei own s eng hs and weaknesses, and a new s anda d is sough ha inhe i s mo e desi able a ibu es, akes in o accoun lessons lea ned om hose p io s anda ds, and achie es mo e widesp ead adop ion. The BIDS p ojec was qui e unique in ha , wi hin he neu oimaging communi y, he e was e ec i ely uni e sal accep ance o ad hoc o ganiza ion o neu oimaging da a beyond he 21 o ma ing o indi idual iles. The bene i s ha we e o be inhe i ed h ough he p ospec o ield-wide ha moniza ion we e he e o e implici ly a ibu ed o BIDS i sel . Clea use cases. One o he main ini ial mo i a ions o de elop BIDS was o allow inges ion o da ase s in o da a eposi o ies a scale wi hou he need o human cu a ion. This no only p o ided a p ac ical p oblem ha de ined he solu ion (as e lec ed by he p ojec p inciples: “adop ion is c ucial”, “don’ ein en he wheel”, and “80/20 ule”), bu also enabled he eam o und he ini ial de elopmen o he s anda d ia he OpenNeu o p ojec . Subsequen ly, BIDS apps p o ided use cases o adop ion by end-use s, by enabling hem o e o lessly p ocess BIDS da ase s using high-quali y so wa e ools. Sol ing a common end use p oblem. S anda dizing da a o ganiza ion was no only bene icial o da a eposi o ies and so wa e pla o ms, bu also o indi idual labs and p ojec s. E en i da a a e ne e sha ed publicly, ha ing a s anda d da a o ganiza ion scheme helps esea che s wi hin a lab g oup wo k oge he , and enables u u e euse o he da a wi hin he lab. The de elopmen o BIDS absol ed indi idual g oups o he need o de elop, documen , and implemen hei own indi idual schemes, and p o ided ools (such as he alida o and con e sion ools) o help wi h his. Mo ing he p ojec away om pu ely “open science” and “da a sha ing” aming by d opping “Open” om he name was c ucial o ein o ce he message ha BIDS was no only o ex e nal sha ing. Low echnical ba ie o en y. The p ac ices adop ed by BIDS do no equi e sophis ica ed ools beyond hose al eady widely in use in mos labo a o ies, enabling apid and widesp ead adop ion. A he same ime, communi y-based e o s a ose o p o ide in oduc o y ma e ials o explain he co e concep s o BIDS o a b oad audience, such as he BIDS S a e Ki 13. I is also wo h acknowledging ha communi y o ums such as Neu oS a s14 ha e played a ole in educa ing use s as well as sha ing bes p ac ices ha all ou side he scope o he s anda d i sel . Ma u i y and he size o he ield. When he BIDS e o s s a ed, human neu oimaging was mo e han 30 yea s old. S anda d pa e ns ac oss expe imen al design and da a ypes had al eady eme ged, and common ile o ma s we e al eady widely adop ed in some sub ields (such as NI TI wi hin he MRI communi y). A he same ime he o e all size o he communi y (se e al housand scien is s, se e al hund ed labs) allowed he eam o ga he de ailed eedback om a signi ican po ion o he communi y and equi ed small o e all alignmen e o . B oad inancial and ins i u ional suppo . BIDS has he bene i o ea ly suppo by INCF (which suppo ed a se o mee ings o he neu oimaging da a sha ing (NIDASH) ask o ce ha included he i s mee ing a S an o d), and majo unding om he Lau a and John A nold Founda ion subsequen ly p o ided suppo o he ea ly de elopmen o BIDS. Since hen, BIDS de elopmen has been suppo ed by g an s om a numbe o ins i u ions (NIMH, NSF, No o 14 h ps://neu os a s.o g/ 13 h ps://bids-s anda d.gi hub.io/bids-s a e -ki / 22 No disk Founda ion, F ench Na ional Resea ch Agency) o a numbe o di e en in es iga o s15, which has b oadened he base o suppo o he p ojec and ensu ed ha i didn’ ely oo hea ily on any one pa icula g an o esea che . In pe son mee ings wi h a di e se se o pa icipan s. Key aspec s o BIDS we e d a ed o e many in-pe son mee ings wi h pa icipan s a eling long and a o a end, and a ious BEPs ha e been s a ed and de eloped spon aneously a con e ences, wo kshops and hacka hons. These in pe son mee ings we e key o b ains o ming he di e en aspec s o he s anda d and cha he wo k ha happened la e asynch onously. Open doo s wi hou “dea h by consensus”. BIDS managed o inely balance ha ing an open s uc u e and lis ening o eedback om many membe s o he neu oimaging communi y, bu a he same ime a oiding ying o please e e yone and c ea ing a s anda d ha was oo lexible o be usable. An example o his was denying ea ly calls o allow he MINC ile o ma in addi ion o NI TI; despi e he echnical supe io i y o MINC, his would ha e made suppo ing he s anda d by any so wa e ool much ha de and diminished i s adop ion. Achie ing his ou come equi ed de leade ship by Ch is Go golewski, which elied hea ily upon his us ed posi ion in he communi y. S umbling poin s o he BIDS p ojec A numbe o challenges ha e a isen du ing he de elopmen o BIDS, which de elope s o o he s anda ds can also possibly lea n om. Delayed adop ion by la ge da abanks. Wi h he excep ion o OpenNeu o, BIDS was no adop ed by any la ge da abases o conso ia du ing i s ea ly de elopmen . Gaining he suppo o a la ge p ospec i e da a sha ing p ojec (such as UK Biobank o Human Connec ome P ojec ) a launch would ha e made i s onge and helped p omo e i . Ins ead, esea che s ha e c ea ed “BIDS-i ied” e sions o hese da ase s, such as he ABCD-BIDS Communi y Collec ion (Feczko e al., 2021). A majo challenge wi h a communi y-d i en p ojec like BIDS is he ela i ely slow pace o de elopmen (due o he need o communi y inpu and con e gence), which can con lic wi h he desi e o la ge p ojec s o sol e hei p oblems quickly wi hou cons ain s om ou side. I would be use ul in he u u e o unding agencies o equi e la ge da a-gene a ion p ojec s o employ communi y s anda ds in o de o enhance he FAIRness o he esul ing da ase s. Challenges o BIDS con e sion. Con e sion o new da ase s in o he BIDS o ma has been, and o en emains, a challenge o many esea che s, which has limi ed e en wide adop ion o he s anda d. Con e e s ha e p oli e a ed16, co e ing mo e use cases bu also po en ially inc easing he bu den on new use s o choose which con e sion ool o adop . Ea lie conce ed e o s on de eloping mo e easily usable con e sion ools could ha e inc eased ea ly adop ion. 16 A cu en lis o BIDS con e e s is a ailable a h ps://bids.neu oimaging.io/bene i s#con e e s 15 Full lis o unding is a ailable a h ps://bids.neu oimaging.io/acknowledgmen s.h ml 23 Lack o a machine- eadable s anda d. Fo many yea s, he e was no common machine- eadable ins an ia ion o he s anda d. This led o misalignmen be ween he s anda d, he alida o , and he documen a ion, and made he implemen a ion o changes in he s anda d di icul and ime-consuming. This has been add essed by he schema iza ion o he s anda d, bu ea lie de elopmen o a machine- eadable s anda d may ha e imp o ed he de elopmen wo k low. Challenges o BEP managemen . While BIDS Ex ension P oposals (BEPs) will emain he main d i e o u u e BIDS de elopmen , hey b ing unique challenges o communi y managemen . Some BEPs may see de elopmen s agna e when aced wi h low de elope a ailabili y o as p ojec equi emen s a e cla i ied. In he bes case, his ela i e do mancy ma ks a clea g ow h poin o he p ojec , as de elope s ealize ha mo e wo k is necessa y o cla i y he BEP scope. In he wo s case, BEPs may be abandoned by he p oposing eam; howe e , i his is eleg aphed app op ia ely, o he communi y membe s can choose o s ep in. O he mo e concep ual challenges include ensu ing he same le el o communi y consul a ion ac oss BEPs. BEP leads who ha e no p e iously pa icipa ed in BIDS de elopmen may no be amilia wi h i s go e nance p ocess. This is pa icula ly conce ning i BEPs a e p eemp i ely yoked o adi ional incen i es, wi h BEP leads commi ing o i m imelines o associa ed publica ions, g an deli e ables, o g adua e deg ee p og ess. Ensu ing ha BEPs main ain he same le el o communi y consul a ion while sus aining he engagemen o domain esea che s mo i a ed o de elop BEPs is an ongoing challenge o BEP managemen . Geog aphical di e si y and inclusi i y. Figu e 6 shows he ins i u ional loca ions o all au ho s on he p esen pape . This map highligh s he ac ha BIDS has ecei ed con ibu ions om a di e se g oup o loca ions ac oss he Uni ed S a es, Wes e n Eu ope, and Aus alia. A he same ime, he e is a no able lack o con ibu ions om esea che s in o he pa s o he wo ld. A goal o he u u e de elopmen o BIDS is o include esea che s om hese pa s o he wo ld ha a e no cu en ly well- ep esen ed in he communi y. 24 Figu e 6. A wo ld map o he ins i u ional loca ions o all coau ho s on he p esen pape . Conclusions Few o us would ha e en isioned in 2015 ha BIDS would be as success ul as i has been, o ha i would wea he he depa u e o i s ini ial ounde so obus ly. This success is a es amen o he sus ained e o s o he la ge numbe o indi iduals who ha e con ibu ed in many di e en ways o he communi y and su ounding ecosys em o ools and da a. We hope ha BIDS can se e as a demons a ion ha communi ies o esea che s can e ec i ely de elop s anda ds ha a e essen ial o enable he e ec i e and FAIR sha ing o esea ch da a. Acknowledgmen s De elopmen o he BIDS S anda d has been suppo ed by he In e na ional Neu oin o ma ics Coo dina ing Facili y, Lau a and John A nold Founda ion, Na ional Ins i u es o Heal h (R24MH114705, R24MH117179, R01MH126699, R24MH117295, P41EB019936, ZIAMH002977, R01MH109682, RF1MH126700, R01EB020740), Na ional Science Founda ion (OAC-1760950, BCS-1734853, CRCNS-1429999, CRCNS-1912266), No o No disk Fonden (NNF20OC0063277), F ench Na ional Resea ch Agency (ANR-19-DATA-0023, ANR 19-DATA-0021), Digi al Eu ope TEF-Heal h (101100700), EU H2020 Vi ual B ain Cloud (826421), Human B ain P ojec (SGA2 785907, SGA3 945539), Eu opean Resea ch Council (Consolida o 683049), Ge man Resea ch Founda ion (SFB 1436/425899996), SFB 1315/327654276, SFB 936/178316478, SFB-TRR 295/424778381), SPP Compu a ional Connec omics (RI 2073/6-1, RI 2073/10-2, RI 2073/9-1), Eu opean Inno a ion Council PHRASE Ho izon (101058240), Be lin Ins i u e o Heal h & Founda ion Cha i é, Johanna Quand Excellence Ini ia i e, ERAPe Med Pa e n-Cog, and he Vi ual Resea ch En i onmen a he Cha i é Be lin – a node o EBRAINS Heal h Da a Cloud. 25 Swedlow, J. R. (2023). OME-Za : a cloud-op imized bioimaging ile o ma wi h in e na ional communi y suppo . bioRxi : The P ep in Se e o Biology. doi: 10.1101/2023.02.17.528834 Mo eau, C. A., Jean-Louis, M., Blai , R., Ma kiewicz, C. J., Tu ne , J. A., Calhoun, V. D., Nichols, T. E., & Pe ne , C. R. (2020). The gene ics-BIDS ex ension: Easing he sea ch o gene ic da a associa ed wi h human b ain imaging. GigaScience,9(10). doi: 10.1093/gigascience/giaa104 Niso, G., Go golewski, K. J., Bock, E., B ooks, T. L., Flandin, G., G am o , A., Henson, R. N., Jas, M., Li ak, V., T Mo eau, J., Oos en eld, R., Scho elen, J.-M., Tadel, F., Wexle , J., & Baille , S. (2018). MEG-BIDS, he b ain imaging da a s uc u e ex ended o magne oencephalog aphy. Scien i ic Da a,5, 180110. Niso, G., Roge s, C., Mo eau, J. T., Chen, L.-Y., Madja , C., Das, S., Bock, E., Tadel, F., E ans, A. C., Jolicoeu , P., & Baille , S. (2016). OMEGA: The Open MEG A chi e. Neu oImage, 124(P B), 1182–1187. No gaa d, M., Ma heson, G. J., Hansen, H. D., Thomas, A., Sea le, G., Rizzo, G., Ve onese, M., Giacomel, A., Yaqub, M., Tonie o, M., Funck, T., Gillman, A., Boni ace, H., Rou ie , A., Dalenbe g, J. R., Be hause , T., Feingold, F., Ma kiewicz, C. J., Go golewski, K. J., … Ganz, M. (2022). PET-BIDS, an ex ension o he b ain imaging da a s uc u e o posi on emission omog aphy. Scien i ic Da a,9(1), 65. Pe ne , C. R., Appelho , S., Go golewski, K. J., Flandin, G., Phillips, C., Delo me, A., & Oos en eld, R. (2019). EEG-BIDS, an ex ension o he b ain imaging da a s uc u e o elec oencephalog aphy. Scien i ic Da a,6(1), 103. Pold ack, R. A., Ba ch, D. M., Mi chell, J. P., Wage , T. D., Wagne , A. D., De lin, J. T., Cumba, C., Koyejo, O., & Milham, M. P. (2013). Towa d open sha ing o ask-based MRI da a: he Open MRI p ojec . F on ie s in Neu oin o ma ics,7, 12. Pold ack, R. A., Feingold, F., F ank, M. J., Gleeson, P., de Hollande , G., Huys, Q. J., Lo e, B. 32 C., Ma kiewicz, C. J., Mo an, R., Ri e , P., Roge s, T. T., Tu ne , B. M., Ya koni, T., Zhan, M., & Cohen, J. D. (2019). The impo ance o s anda ds o sha ing o compu a ional models and da a. Compu a ional B ain & Beha io ,2(3-4), 229–232. Robbins, K., T uong, D., Jones, A., Callanan, I., & Makeig, S. (2022). Building FAIR Func ionali y: Anno a ing E en s in Time Se ies Da a Using Hie a chical E en Desc ip o s (HED). Neu oin o ma ics,20(2), 463–481. Robb, R. A., Hanson, D. P., Ka woski, R. A., La son, A. G., Wo kman, E. L., & S acy, M. C. (1989). Analyze: a comp ehensi e, ope a o -in e ac i e so wa e package o mul idimensional medical image display and analysis. Compu e ized Medical Imaging and G aphics: The O icial Jou nal o he Compu e ized Medical Imaging Socie y,13(6), 433–454. Roge s, C., Wickenheise , A., Fesselie , L., Callega o, J., Madja , C., Das, S., & E ans, A. C. (2022). EEG2BIDS Wiza d: a lexible open-sou ce c oss-pla o m ool o anonymiza ion and BIDS s anda diza ion. 14 h Annual Mee ing o he O ganiza ion o Human B ain Mapping. Sanz Leon, P., Knock, S. A., Woodman, M. M., Domide, L., Me smann, J., McIn osh, A. R., & Ji sa, V. (2013). The Vi ual B ain: a simula o o p ima e b ain ne wo k dynamics. F on ie s in Neu oin o ma ics,7, 10. Schi ne , M., Domide, L., Pe dikis, D., T iebko n, P., S e ano ski, L., Pai, R., P odan, P., Valean, B., Palme , J., Lang o d, C., Blickensdö e , A., an de Vlag, M., Diaz-Pie , S., Peyse , A., Klijn, W., Plei e , D., Nahm, A., Schmid, O., Woodman, M., … Ri e , P. (2022). B ain simula ion as a cloud se ice: The Vi ual B ain on EBRAINS. Neu oImage,251, 118973. She i , T., Rioux, P., Rousseau, M.-E., Kassis, N., Beck, N., Adala , R., Das, S., Gla a d, T., & E ans, A. C. (2014). CBRAIN: a web-based, dis ibu ed compu ing pla o m o collabo a i e neu oimaging esea ch. F on ie s in Neu oin o ma ics,8, 54. Tee e s, J. L., God ey, K., Young, R., Dang, C., F iedsam, C., Wa k, B., Asa i, H., Pe on, S., Li, 33 N., Pey ache, A., Deniso , G., Siegle, J. H., Olsen, S. R., Ma in, C., Chun, M., T ipa hy, S., Blanche, T. J., Ha is, K., Buzsáki, G., … Somme , F. T. (2015). Neu oda a Wi hou Bo de s: C ea ing a Common Da a Fo ma o Neu ophysiology. Neu on,88(4), 629–634. Van Essen, D. C., Smi h, S. M., Ba ch, D. M., Beh ens, T. E. J., Yacoub, E., Ugu bil, K., & WU-Minn HCP Conso ium. (2013). The WU-Minn Human Connec ome P ojec : an o e iew. Neu oImage,80, 62–79. Van Ho n, J. D., & Gazzaniga, M. S. (2013). Why sha e da a? Lessons lea ned om he MRIDC. Neu oImage,82, 677–682. Wilkinson, M. D., Dumon ie , M., Aalbe sbe g, I. J. J., Apple on, G., Ax on, M., Baak, A., Blombe g, N., Boi en, J.-W., da Sil a San os, L. B., Bou ne, P. E., Bouwman, J., B ookes, A. J., Cla k, T., C osas, M., Dillo, I., Dumon, O., Edmunds, S., E elo, C. T., Finke s, R., … Mons, B. (2016). The FAIR Guiding P inciples o scien i ic da a managemen and s ewa dship. Scien i ic Da a,3, 160018. Ya koni, T., Ma kiewicz, C. J., de la Vega, A., Go golewski, K. J., Salo, T., Halchenko, Y. O., McNama a, Q., DeS asio, K., Poline, J.-B., Pe o , D., Hayo -Sasson, V., Nielson, D. M., Ca lin, J., Kia , G., Whi ake , K., DuP e, E., Wagne , A., Ti ell, L. S., Jas, M., … Blai , R. (2019). PyBIDS: Py hon ools o BIDS da ase s. Jou nal o Open Sou ce So wa e,4(40). doi: 10.21105/joss.01294 34