Open and ep oducible esea ch in MSK imaging: Why i ma e s and how o implemen i wi h he guidelines o he ORMIR communi y 1
JBMRPlus,2025, 1–9
doi: DOI HERE
Ad ance Access Publica ion Da e: Day Mon h Yea
Pe spec i e Pape
Open and ep oducible esea ch in musculoskele al
imaging: Why i ma e s and how o implemen i
wi h he guidelines o he ORMIR communi y
Se ena Bona e i ,1,∗Moj aba Ba zega i ,2Melissa Be e s ,3,4,5 S e en Boyd ,6,7
And ew J. Bu gha d ,8Donnie Came on ,9F ancesco Chiumen o ,10
Gianluigi C imi11, Ge a d Degenha ,12 Pholpa Du ongbhan ,13
Michelle Alejand a Espinosa He nandez ,14,13 Giulia F a e igo 11,
Ali Ghasem-Zadeh ,15 Lo enzo G assi ,16 Jukka Hi asniemi ,17
Mahdi Hosseini aba abaei ,18 Gianluca Io i ,19 Joe i Kok ,5Michael Kuczynski ,20,7
YoungJun Lee ,21 Cecilia Libe a i 22, Sa ah Manske ,6,7 Ma McCo mick ,23
Ma ia Monzon ,24 Ma ino Pani,25 Simone Poncioni ,26 Jilmen Quin iens 22,
Sabine R¨aube ,27 Paul Ri sche ,28 Al onso Da io San ama ia11,
F ancesco San ini ,27 Fabio Sa o ,29 En ico Schileo11, Vincen S adelmann ,30
Ka h yn S. S ok ,13 Rachel Su owiec ,21 Ful ia Taddei11, Ja ed Vico y,31
Ma hias Walle ,7Ma iska Wesseling ,32 Danielle Whi ie ,20,7 Be ina Willie 33,
Andy Kin On Wong 34,35 and Dˇzenan Zuki´c 31
1Independen Resea che , Swi ze land, 2Depa men o Chemical Enginee ing and Chemis y, Eindho en Uni e si y o Technology (TU/e), The Ne he lands,
3Depa men o In e nal Medicine, VieCu i Medical Cen e , The Ne he lands, 4NUTRIM Ins i u e o Nu i ion and T ansla ional Resea ch In Me abolism,
Maas ich Uni e si y, The Ne he lands, 5Depa men o Biomedical Enginee ing, Eindho en Uni e si y o Technology, The Ne he lands, 6Depa men o
Radiology, Cumming School o Medicine, Uni e si y o Calga y, Canada, 7McCaig Ins i u e o Bone and Join Heal h, Uni e si y o Calga y, Albe a,
Canada, 8Depa men o Radiology & Biomedical Imaging, Uni e si y o Cali o nia, San F ancisco, CA, USA, 9Depa men o Medical Imaging, Radboud
Uni e si y Medical Cen e , The Ne he lands, 10School o Elec onic Enginee ing, Dublin Ci y Uni e si y (DCU), I eland, 11Bioenginee ing and Compu ing
Labo a o y, IRCCS Is i u o O opedico Rizzoli, I aly, 12Co e Faicili y Mic oCT, Uni e si y Clinic o Radiology, Medical Uni e si y Innsb uck, Aus ia,
13Depa men o Biomedical Enginee ing, The Uni e si y o Melbou ne, Vic o ia, Aus alia, 14Rehabili a ion Sciences Ins i u e, Uni e si y o To on o,
ON, Canada, 15Depa men s o Endoc inology and Medicine, Aus in Heal h, The Uni e si y o Melbou ne, Vic o ia, Aus alia, 16Depa men o Biomedical
Enginee ing, Lund Uni e si y, Sweden, 17Depa men o Radiology & Nuclea Medicine, E asmus MC Uni e si y Medical Cen e Ro e dam, The Ne he lands,
18Depa men o Bioenginee ing, Facul y o Enginee ing, McGill Uni e si y, Quebec, Canada, 19Cen e o Pho on Science, Paul Sche e Ins i u PSI,
Swi ze land, 20Depa men o Cell Biology and Ana omy, Cumming School o Medicine, Uni e si y o Calga y, Albe a, Canada, 21Weldon School o
Biomedical Enginee ing, Pu due Uni e si y, IN, USA, 22Biomechanics Sec ion, Depa men o Mechanical Enginee ing, KU Leu en, Belgium, 23Fideus Labs,
NC, USA, 24Depa men o Heal h Science and echnology, ETH Zu ich, Zu ich, Swi ze land, 25School o Elec ical and Mechanical Enginee ing, Uni e si y
o Po smou h, Hampshi e, Uni ed Kingdom, 26ARTORG Cen e o Biomedical Enginee ing Resea ch, Uni e si y o Be n, Swi ze land, 27Basel Muscle MRI,
Depa men o Biomedical Enginee ing, Uni e si y o Basel, Basel, Swi ze land, 28Depa men o Spo , Exe cise and Heal h, Uni e si y o Basel, Swi ze land,
29Depa men o Biomedical Sciences, Uni e si y o Pado a, I aly, 30Depa men o Resea ch and De elopmen , Schul hess Klinik, Swi ze land, 31Medical
Compu ing, Ki wa e, Inc., NC, USA, 32Depa men o O hopedics and Spo s Medicine, E asmus Medical Cen e , The Ne he lands, 33Facul y o Den al
Medicine and O al Heal h Sciences, McGill Uni e si y and Sh ine s Hospi al o Child en-Canada, Quebec, Canada, 34Join Depa men o Medical Imaging;
Sch oede A h i is Ins i u e, Uni e si y Heal h Ne wo k, ON, Canada and 35Dalla Lana School o Public Heal h, Uni e si y o To on o, ON, Canada
∗Co esponding au ho : se ena.bona e i. esea[email p o ec ed]
Abs ac
The Open and Rep oducible Musculoskele al Imaging Resea ch (ORMIR) communi y is a scien i ic com-
muni y dedica ed o p omo ing openness and ep oducibili y in musculoskele al imaging esea ch. In his
pe spec i e pape , we ou line he mo i a ions o conduc ing anspa en esea ch and p o ide p ac ical
guidelines o implemen i . We s a wi h de ining open and ep oducible esea ch and desc ibing he bene-
i s and challenges o wo king anspa en ly. Nex , we ede ine he ou pu s o a compu a ional esea ch
s udy as a combina ion o da a, code, and a publica ion, ecommend a olde and ile s uc u e ha
e lec s hese h ee s udy ou comes, and desc ibe how o main ain and upda e such s uc u e du ing he
s udy and a s udy publica ion. Finally, we emphasize ha wo king in an open and ep oducible man-
ne is a lea ning p ocess and he bes way o acqui e he necessa y compe encies is simply o s a .
Lay summa y: The ORMIR communi y p omo es openness and ep oducibili y in musculoskele al ima-
ging esea ch. In his pe spec i e pape , we explain why anspa ency ma e s and ecommend how o
conduc a compu a ional s udy in an open and ep oducible manne ocusing on i s h ee ou pu s: da a,
code, and publica ion. Finally, we highligh ha he bes way o lea n hese p ac ices is simply o s a .
Keywo ds: Open esea ch, ep oducible esea ch, da a sha ing, open-sou ce code, compu a ional na a i es,
guidelines, license, ORMIR
In oduc ion
In 2012, Panagio opoulou e al. published an a icle i led “Wha
makes an accu a e and eliable subjec -speci ic ini e elemen
model? A case s udy o an elephan emu ” (1). In his wo k,
hey sough “ o cons uc he i s eliable ini e elemen model
o he emu o an adul Asian elephan ” om compu ed omo-
g aphy images. As pa o he s udy, he au ho s conduc ed
con e gence analyses o he size, ype, and numbe o elemen s
used in he geome y disc e iza ion o he emu ana omy. In
he ollowing yea s, he pape ecei ed 33 ci a ions1. Some o
hese publica ions e e ed o he me hodology and indings o
his s udy in hei in oduc ion o discussion, while o he publi-
ca ions buil upon he s udy me hodology and indings. Among
he la e , in 2018 B assey e al. published a s udy in which
hey used he same size, ype, and numbe o elemen s as in
Panagio opoulou’s publica ion, mo i a ing ha “p e ious esea-
ch has ound such meshes pe o m well” and “s ain alues ha e
been ound o con e ge when elemen numbe s exceed 200,000”
(2). While building upon p e ious wo k is ypical in esea ch, in
his case he e is an unexpec ed ou come. In 2014— ou yea s
be o e B assey’s publica ion—Panagio opoulou’s pape had been
sel - e ac ed because he au ho s “unco e ed p oblems wi h he
me hods [...] and he e o e some o he inal da a” (3).
This s o y highligh s h ee impo an conside a ions. Fi s ,
Panagio opoulou e al. dese e high p aise o hei scien i ic in e-
g i y. Second, wha occu ed o B assey e al. could happen o
any esea che who builds a s udy on p e ious indings. Thi d, we
esea che s can be inspi ed o ask undamen al ques ions abou
some p ac ical needs ha we ace in ou daily wo k: How can we
and ou collabo a o s c ea e and publish e i iable me hodologies
and indings ha o he s can build upon? How can we a oid buil-
ding ou s udies on inco ec me hods and esul s? And how can
we check he co ec ness o me hodologies and esul s when we
e iew a pape and assess i s scien i ic alue? The answe o hese
ques ions is open and ep oducible esea ch.
Open and ep oducible esea ch: De ini ions,
bene i s, and challenges
Openness e e s o “an app oach o esea ch based on open coo-
pe a i e wo k ha emphasizes he sha ing o knowledge, esul s
and ools as ea ly and widely as possible” (4). Rep oducibili y—
in ended he e as compu a ional ep oducibili y—is he abili y o
esea che s o duplica e he esul s o a p e ious s udy using he
“same inpu da a, compu a ional s eps, me hods, and condi ions
o analysis” (5). Toge he , openness and ep oducibili y cons i u e
away o wo king whose ul ima e goal is o imp o e he eliabili y
and e iciency o scien i ic p og ess (6; 7).
The scien i ic ad an ages o open and ep oducible esea ch
a e se e al and include e alua ing he co ec ness o scien i-
ic claims (8), building on p e ious wo k wi h con idence and
e iciency (9)—by eusing and ex ending exis ing compu a io-
nal pipelines—and compa ing and alida ing new me hods and
algo i hms agains exis ing ones, wi hou ha ing o eimplemen
hem (10). These scien i ic bene i s a e accompanied by academic
ad an ages. Publishing open da a can inc ease he numbe o ci a-
ions by app oxima ely 70%, ega dless o jou nal impac ac o ,
1Re ie ed om Google Schola : h ps://schola .google.com.
Accessed: July 21, 2025.
au ho coun y o o igin, and da e o publica ion (11). Simila ly,
making code a ailable online can inc ease ci a ions by a ac o o
h ee (12). Fu he mo e, esea ch labo a o ies can g ea ly bene i
om adop ing open and ep oducible p ac ices. Such a anspa-
en me hodological app oach os e s an in insic “discipline” ha
suppo s he con inui y o main enance and ex ension o compu a-
ional ools c ea ed in he labs, which would o he wise isk a sho
li espan as hey a e usually de eloped by successi e and dis inc
gene a ions o PhD s uden s and pos docs (6).
Despi e he a o emen ioned ad an ages, wo king in an open
and ep oducible manne has been challenging. Un il a ew yea s
ago, guidelines we e sca ce, compu a ional ools we e no ully
de eloped, and pe manen , la ge, and ee eposi o ies o da a
and code we e una ailable (13). In addi ion, many scien is s we e
conce ned abou no ecei ing p ope ecogni ion, ha ing o allo-
ca e addi ional ime and e o o p epa e da a and code o elease,
and p o iding ad an ages o compe ing da ase s o so wa e when
eleasing seminal wo k (14). Howe e , he mos c i ical limi a ion
o sha ing da a and code may be ela ed o he way we concei e
he ou come o compu a ional s udies.
Wha is he ou come o a compu a ional esea ch
s udy?
T adi ionally, he main ou come o a compu a ional esea ch s udy
is a scien i ic pape . As we know, pape s a e concise epo s limi-
ed by a de ined numbe o wo ds ( ypically 3,000–6,000) and a
es ic ed numbe o igu es and ables (usually 3–6). To mee
hese cons ain s, only a selec ion o he me hods and esul s
de eloped and ob ained du ing a s udy a e p esen ed in a pape .
C ucial implemen a ion de ails (e.g., compu a ional ini ializa ions
and pa ame e s) a e o en omi ed (12), as well as in e media e
o indi idual indings, because esul s a e usually p esen ed as
agg ega ed s a is ics, such as means and s anda d de ia ions. In
his con ex , a scien i ic pape does no ep esen “ he schola ship
i sel , i ’s me ely schola ship ad e isemen ”, as Donoho amously
w o e in 2009 (6). In o he wo ds, a scien i ic pape is jus he ip
o he icebe g ep esen ing esea ch e o s (15). The ac ual wo k—
ha is, he da a and code— emains hidden unde wa e (Figu e 1,
le ).
In open and ep oducible esea ch, da a and code eme ge o
be an in eg a i e pa o he ou come o a compu a ional s udy.
Toge he wi h publica ions, hey cons i u e he h ee essen ial,
in e locked pieces o he puzzle necessa y o unde s and and ep o-
duce he en i e s udy (Figu e 1, igh ). F om he aw da a o he
inpu , h ough he code, esea che s can ec ea e he esul s p e-
sen ed in he publica ion, and subsequen ly euse and ex end he
wo k low o new s udies. Howe e , o his o be easible, da a
and code should be clea ly s uc u ed and documen ed (16).
Mo e ecen ly, nume ous guidelines ha e been a ailable on
how o publish compu a ional s udies in an open and ep odu-
cible manne while enabling a ibu ion and acili a ing sha ing
(8; 9; 17; 18; 19). Mo eo e , esea che s in many ields ha e
o med scien i ic communi ies o es ablish consensus-based s an-
da ds and de elop compu a ional ools speci ic o hei disciplines
(20; 21; 22).
2
©
The Au ho 2025. Published by Ox o d Uni e si y P ess. All igh s ese ed. Fo pe missions, please e-mail: jou nals.p[email p o ec ed]
Open and ep oducible esea ch in MSK imaging: Why i ma e s and how o implemen i wi h he guidelines o he ORMIR communi y 3
Publica ions
Da a and Code Publica ions
Code
Da a
Fig. 1: Rep esen a ion o esea ch ou comes. T adi ionally, publica ions a e he main esea ch ou come. They cons i u e he sha ed,
isible ip o an icebe g, whe eas da a and code emain hidden in he lab (le ; modi ied om h ps://s odden.ne /AMP2011/slides/
2011-amp- ep oducible- esea ch.pp x.pd ). Con e sely, in open and ep oducible esea ch da a, code, and publica ion cons i u e he
h ee in e locking pieces o he puzzle and a e essen ial o each o he o ep esen he en i e s udy ( igh ).
ORMIR guidelines o open and ep oducible
compu a ional s udies in MSK imaging esea ch
In musculoskele al (MSK) imaging, he Open and Rep oduci-
ble Musculoskele al Imaging Resea ch (ORMIR) communi y was
ounded in 2020, wi h he aims o c ea ing open, ep oducible,
well- es ed, and well-documen ed code o analyze MSK images;
s anda dizing da a acquisi ion and managemen o a o da a
sha ing and algo i hm compa ison; and p omo ing a cul u e o
openness and ep oducibili y in musculoskele al imaging esea-
ch o a as e ad ancemen o he ield (23). The communi y
is cu en ly composed o app oxima ely 70 in e na ional esea-
che s om academia and indus y a a ious ca ee s ages, who
ha e been ac i e in c ea ing guidelines, empla es, and so wa e o
manage and analyze MSK images—see www.o mi .o g.
In he nex wo sec ions o his a icle, we will p esen he
ORMIR guidelines on how o c ea e an open and ep oducible
compu a ional s udy in MSK imaging. These guidelines a e based
on he p inciple ha open and ep oducible esea ch cons i u es
a me hodological app oach ha should be implemen ed h ough-
ou an en i e s udy, a he han being conside ed as a ime- and
esou ce-in ensi e ask a publica ion. The e o e, i s , we will
ocus on how o conduc a compu a ional MSK esea ch s udy
o an open and ep oducible publica ion. Then, we will p o ide
sugges ions on how o in eg a e and sha e he h ee s udy ou co-
mes a publica ion. All ecommenda ions a e summa ized in Fig.2
and can be adap ed o he speci ic needs o any compu a ional
s udy.
Conduc ing a compu a ional MSK esea ch s udy o
an open and ep oducible publica ion
Begin by o ganizing you esea ch p ojec in o a olde s uc u e
simila o he one in Fig. 3. You can download he olde s uc u e
om Zenodo (www.doi.o g/10.5281/zenodo.17206440) o imple-
men i manually. S a wi h a main olde and name i a e he
s udy. Wi hin he main olde , c ea e h ee addi ional olde s, each
o one o he h ee s udy ou comes: da a,code, and publica ion.
The s uc u e and con en o each olde is desc ibed in mo e de ail
in he ollowing sec ions.
Da a
Wi hin he da a olde , add ou olde s: aw,de i ed, and
esul s o da a a di e en s ages o he ep oducible wo k low,
and da a docs o documen a ion.
In he olde aw, s o e he o iginal da a o he s udy, ha is,
he image iles—usually in DICOM o p op ie a y o ma —and
demog aphic o clinical in o ma ion o he subjec s—commonly
in abula o ma . These da a cons i u e he inpu o you wo k-
low. Do no modi y hem ei he ia compu a ional o manual
in e en ion o ensu e he ep oducibili y o he whole s udy
(8; 24).
In he olde de i ed, sa e he in e media e da a gene a ed
du ing he compu a ional wo k low. Fo he in e nal s uc u e o
his olde , ollow he speci ica ions de ined by he ORMIR Muscu-
loskele al Imaging Da a S uc u e (ORMIR-MIDS) (25), which
ex ends he B ain Imaging Da a S uc u e (BIDS) p inciples (26)
speci ically o MSK imaging. Reo ganize he image iles in aw
in o a hie a chy o nes ed olde s a anged by subjec , session,
and imaging modali y using he ORMIR-MIDS Py hon package,
which also con e s he images om he Digi al Imaging and
Communica ions in Medicine (DICOM) o ma (27) o he Neu-
oimaging In o ma ics Technology Ini ia i e (NI TI) o ma (28)
(Fig. 3, igh ). Du ing he s udy, s o e he in e media e ou -
pu s o he compu a ional wo k low—such as segmen a ion masks,
mo phological measu emen s, and ini e elemen analyses—in his
olde s uc u e. Sa ing hese in e media e ou pu s will acili a e
debugging and po en ial euse o pa ial indings (9).
4Se ena Bona e i e al.
Conduc ing a compu a ional MSK esea ch s udy
o an open and ep oducible publica ion
Publishing a compu a ional MSK esea ch s udy
as open and ep oducible
Da a
•
C ea e 4 olde s: aw, de i ed, esul s, and da a_docs
•Ne e modi y he da a in aw
•Use he ORMIR-MIDS s uc u e o he in e media e
da a
•The da a in esul s will appea in he publica ion
•Documen he who/wha /when/whe e/why/how
•
Use open ile o ma s and meaning ul ile names
•
Ve i y sha ing easibili y:
•Conside consen and anonymiza ion
•Consul he ORMIR Da a Sha ing Guidelines
•
Selec a eposi o y p o iding DOI, e sion con ol, and
emba go i needed
•
A oid links o cloud se ices o pe sonal websi es
•
Choose an ins i u ional o a C ea i e Commons license
Code
•
C ea e 4 olde s: s c, no ebooks, ba ch_sc ip s, and
code_docs
•Add he Py hon modules o s c
•Use Jupy e No ebooks o analyze abula da a,
gene a e g aphs o publica ions, and c ea e demos–
use he ORMIR empla e
•Add he bash sc ip s o ba ch_sc ip s
•Documen code using Sphynx, Read he Docs o
Jupy e Book
•W i e code o humans: use a iables wi h desc ip i e names
and ollow s yle con en ions
•
Use e sion con ol (Gi Hub)
•
Change he Gi Hub eposi o y isibili y o public
•
Use Binde o sha e no ebooks in e ac i ely
•
Hos code documen a ion on Gi Hub o Read he Docs
•
Sync he code wi h a eposi o y p o iding DOI o ci a ions
•
Choose an open-sou ce license
Publica ion
•
W i e he manusc ip in La ex, Ma kdown, o Google Docs
•
Add links o code and da a eposi o ies o he pape
•
Add links o compu a ional wo k lows o he g aphs
•
Upload o a p ep in se e
•
Submi o an open access jou nal
Takeaways
•
I is a lea ning p ocess, and you will lea n by doing
•
In case o doub s, ask he ORMIR communi y
•
I does no ha e o be pe ec . Jus s a !
•
O he s can now build on you esea ch
•
You ha e con ibu ed o a mo e anspa en and as e
ad ancemen o MSK imaging esea ch
Fig. 2: Summa y o bes p ac ices o open and ep oducible compu a ional s udies in musculoskele al imaging esea ch.
In he olde esul s, sa e he inal ou pu o he compu a-
ional wo k low, including abula iles epo ing mo phological,
densi ome ic, mechanical, o s a is ical calcula ions ac oss subje-
c s. These a e he da a ha you will p esen in he Resul s sec ion
o he publica ion.
Finally, in he olde da a docs, add he da a documen a ion,
such as acquisi ion p o ocol, e hical app o al, and any me ada a—
ha is, in o ma ion abou he da a. As cu en ly ORMIR does no
p o ide speci ic guidelines o da a documen a ion, simply collec
all he easonable in o ma ion ha speci y he “who/wha /whe-
n/whe e/why/how” o you da ase (13), so ha o he esea che s
can unde s and and euse i .
A e e y s age o he compu a ional wo k low, use open and
widely suppo ed ile o ma s (24). Fo imaging da a, p e e NI TI
o o he open s anda ds o scanne -speci ic p op ie a y o ma s.
Fo abula da a, use comma-sepa a ed alues (.cs ) o ab-
sepa a ed alues (. s ) o e p op ie a y o ma s, such as Excel.
Finally, gi e iles meaning ul names ha clea ly indica e hei
con en (24).
Code
In he olde code, c ea e he ou olde s s c,no ebooks,
ba ch sc ip s, and code docs—in la ge p ojec s add he addi-
ional olde es s o au oma ed uni es s (29).
In he olde s c (abb e ia ion o sou ce), add he co e algo-
i hms o he s udy. W i e code in he open-sou ce p og amming
language Py hon, which has gained popula i y in medical imaging
o he a ailabili y o gene al-pu pose image p ocessing packages—
such as ITK (30) and SimpleITK (31)—as well as deep lea ning
and machine lea ning packages—including PyTo ch (32), Tenso -
Flow (33), and Sciki -lea n (34). Wi hin s c, o ganize you code
in o Py hon modules, ha is, .py iles. Fo la ge p ojec s, g oup
he modules in o a Py hon package and change he olde name o
he package name, as equi ed by he packaging ools (35).
In he olde no ebooks, add he s udy’s Jupy e No ebooks,
which a e he ideal o ma o in eg a ing code wi h na a i e,
equa ions, and isualiza ions o e ec i ely documen ep oducible
wo k lows (9). Download he Jupy e No ebook empla e c ea ed
by he ORMIR communi y om Zenodo (www.doi.o g/10.5281/
zenodo.17206440) and adap i o you s udy. The empla e has
a clea s uc u e composed o aims, compu a ional s eps, ou pu s,
and dependencies— ha is, a p in o he ha dwa e cha ac e is ics
and Py hon package e sions (9). Use no ebooks o analyze abula
da a, gene a e igu es o he publica ion, collec he main unc ion
calls om he modules in s c in o a compu a ional wo k low wi h
na a i e, and c ea e examples o demos o you p ojec .
In he olde ba ch sc ip s, sa e he bash iles. These iles a e
he sc ip s ha call he algo i hms om he olde s c and ha
you use o la ge-scale compu a ions on se e s.
In all cases, w i e code “ o humans”, ha is, make i eadable
a he han unnecessa ily complex (36). To c ea e clea code, use
a iables wi h desc ip i e names ha e lec hei ole in he com-
pu a ion (36) and adhe e o he Py hon coding con en ion PEP8
(37). Mo eo e , modula ize code in o unc ions—o classes— ha
ep esen uni s o ope a ions (24; 9). Documen each unc ion
ollowing he NumPy S yle (38) o Google Py hon S yle (39)—
which clea ly desc ibe unc ionali y, inpu s, and ou pu s—and
add inline commen s o explain non-ob ious s eps and cla i y he
unde lying logic. Finally, synch onize he code o a eposi o y like
Gi Hub (h ps://gi hub.com/). Such eposi o ies p o ide e sion
con ol o p ese e he his o y o changes and e e o p e ious
Open and ep oducible esea ch in MSK imaging: Why i ma e s and how o implemen i wi h he guidelines o he ORMIR communi y 5
s udy_name
code
s c
no ebooks
p ep ocessing.py
p ep ocess_images.ipynb
segmen a ion.py
segmen _images.ipynb
da a
aw
de i ed
902145989_001.dcm
sub-01
798346781_001.dcm
sub-02
ba ch_sc ip s
un_segmen a ion.sh
un_p ep ocessing.sh
esul s
mo phology.cs
s a is ics.cs
publica ion
code_docs
sphynx/ ead hedocs/jupy e book
pape . ex
ses-01
m -ana
c
sub-01_ses-01_T1w.json
sub-01_ses-01_c .json
sub-01_ses-01_T1w.nii.gz
sub-01_ses-01_c .nii.gz
sub-01_ses-01_seg.nii.gz
sub-01_ses-01_seg.nii.gz
sub-01_ses-01_mo ph. s
ses-02
ORMIR-MIDS
da a_docs
acquisi ion.pd
Fig. 3: Folde and ile s uc u e o a compu a ional esea ch s udy o an open and ep oducible publica ion.
code in case o e o s, and hey a e con enien o coo dina ing
p og amming asks in collabo a i e so wa e de elopmen (17).
Finally, in he olde code docs (o simply docs) c ea e code
documen a ion o de elope s and use s, especially i you a e buil-
ding a la ge p ojec (40). In he documen a ion o de elope s,
ca e ully explain he echnical de ails o he code. C ea e his docu-
men a ion using Sphynx (41) o Read he Docs (42) as hey a e
compa ible wi h he NumPy and Google Py hon s yles. In he
documen a ion o use s, explain in non echnical language how
o ins all and execu e you code and p o ide examples o demos.
W i e his documen a ion using Jupy e Book (43).
Publica ion
In he las olde publica ion, s o e all he ma e ial ela i e o
he scien i ic pape . W i e he manusc ip in a o ma ha pe mi s
e sion con ol, such as LaTeX o Ma kdown, and op o online
ools ha allow acking changes and collabo a i e edi ing, such
as O e lea (www.o e lea .com/) o Google Docs (h ps://docs.
google.com) (24).
Publishing a compu a ional MSK esea ch s udy as
open and ep oducible
A publica ion, he p ima y ask is o decide how and whe e o
sha e he h ee s udy ou comes. He e a e he de ails o each o
hem.
Da a
Sha ing s udy da a equi es a en ion o wo main aspec s: asses-
sing he easibili y o sha ing and choosing a da a eposi o y
(44; 45).
The easibili y o sha ing depends on legal and e hical ac o s,
including whe he he s udy pa icipan s signed an app op ia e
consen o m, he le el o da a anonymiza ion o deiden i ica ion
equi ed, and he p i acy legisla ion o you ins i u ion and coun-
y. These ac o s can p eclude sha ing some o all he con en o
he olde s aw,de i ed, and da a docs. Howe e , you can always
sha e he inal agg ega ed da a o you s udy— ha is, he con en
o he olde esul s. Fo u he de ails, consul he ORMIR Da a
Sha ing Guidelines (46).
To da e, only a ew dedica ed MSK eposi o ies exis whe e
you can sha e you da a, such as he Uni e sal Musculoskele al
Ul asound Reposi o y (UMUD) (47). I you da a does no i
he sco e o such eposi o ies, sha e he da a in you ins i u ion

6Se ena Bona e i e al.
eposi o y, ollowing hei egula ion. O he wise, use gene al pu -
pose eposi o ies such as Zenodo (h ps://zenodo.o g)—whe e
ORMIR is p esen as a communi y label—o Figsha e (h ps:
// igsha e.com). Bo h hese eposi o ies p o ide a digi al objec
iden i ie (DOI) o ensu e pe manen a ailabili y and enable ci a-
ion, e sioning o da a, and an emba go pe iod i you canno
immedia ely sha e you da a. A oid links o pe sonal websi es o
cloud s o age, because hese links end o expi e o be dele ed
(48; 49).
Code
To make you code open, simply change he Gi Hub eposi o y
isibili y om p i a e o public. Comple e he README.md ile
o he eposi o y using he empla e p o ided by he ORMIR
communi y (www.doi.o g/10.5281/zenodo.17206440). Whene e
possible, sha e he no ebooks in an in e ac i e and ep oducible
cloud en i onmen , such as Binde (h ps://mybinde .o g) (50).
Hos he code documen a ion c ea ed wi h Sphynx o Jupy e
Book on Gi Hub Pages (51; 52) o he hos ing se ice p o ided by
Read he Docs. Finally, synch onize he Gi Hub eposi o y wi h
Zenodo (53), Fighsha e (54), o So wa e He i age (55) o ob ain
a DOI ha enables ep oducibili y and ci a ions.
Publica ion
Add he links o da a and code eposi o ies o he manusc ip . I
possible, add he speci ic links o he da a and code used o c ea e
he g aphs in hei cap ion o enable eade s o “ ecalcula e he
igu e om all i s da a, pa ame e s, and p og ams” (56). Upload
he pape o a p ep in a chi e, such as a Xi (www.a xi .o g),
bioRxi (www.bio xi .o g), o medRxi (www.med xi .o g), and
hen submi i o a jou nal ha suppo s open access. In he no
oo dis an u u e, we may be able o w i e ully execu able and
in e ac i e pape s in which eade s can e un he desc ibed expe-
imen s (13). This ision is suppo ed by exis ing pilo p ojec s,
whe e Jupy e No ebooks a e submi ed oge he wi h con e ence
abs ac s (57; 58).
Licenses
Finally, add a license o you da a, code, and publica ion o speci y
how you wan hem o be eused and modi ied. Wi hou a license,
you would e ain all he igh s and o he esea che s would no
be allowed o use you ma e ial (24). I a ailable, use he licenses
ecommended by you ins i u ion. O he wise, he e a e he ORMIR
guidelines.
Fo da a and publica ion, use he C ea i e Commons (CC)
license, whose ea u es—a ibu ion (BY), Non Comme cial use
(NC), Sha e-Alike (SA), and No De i a i e (ND)—can be com-
bined o c ea e a cus omized license. C ea i e Commons also
p o ides he CC0 license, which pe mi s eleasing ma e ial o
he public domain wi hou e aining any copy igh (h ps://
c ea i ecommons.o g/sha e-you -wo k/cclicenses/).
Fo code, wo main ypes o open-sou ce licenses exis :
es ic i e—which ex end openness o de i ed wo k (e.g., GNU
licenses)—and pe missi e, which do no equi e openness o
de i ed wo k (e.g., MIT, BSD, o Apache) Fo a comple e lis
o licenses wi h hei pe missions and limi a ions, see h ps:
//choosealicense.com/appendix. Choose a license ha is com-
pa ible wi h he licenses o he compu a ional ools you used in
you s udy and ha you conside he mos app op ia e o he
u u e o you p ojec (18).
Jus s a !
Open and ep oducible esea ch is a way o wo king ha sup-
po s ou needs as esea che s o assess scien i ic claims, build on
p e ious eliable wo k, and con ibu e o as e ad ancemen o
ou esea ch ield. In his pe spec i e pape , we p oposed conc e e
guidelines c ea ed by he ORMIR communi y on how o wo k in
an open and ep oducible manne . Adop ing his me hodological
app oach is a lea ning p ocess ha is mos e ec i e when lea ned
by doing. I you ha e doub s abou some guidelines o ools, con-
ac he ORMIR communi y. Do no aspi e o pe ec ion om he
e y beginning. Jus s a ! You e o s will p o e aluable because
o he esea che s will be able o deeply unde s and and build on
you esea ch, and you will ha e con ibu ed o a mo e anspa en
and as e ad ancemen o MSK imaging esea ch.
Compe ing in e es s
The au ho s ha e no compe ing in e es s o decla ed.
Au ho con ibu ions s a emen
Au ho ship was o e ed o any indi idual who has con ibu ed
o he de elopmen o he ORMIR communi y and i s guide-
lines. Au ho s we e asked o deno e hei con ibu ions o he
p ojec acco ding o he Con ibu o Roles Taxonomy (CRediT).
Role abb e ia ions. C: Concep ualiza ion, DC: Da a cu a ion o
ini ial use and la e e-use, FA: Funding acquisi ion, M: Me hodo-
logy, PA: P ojec adminis a ion, R: Resou ces, S: So wa e, SU:
Supe ision, V: Valida ion, W: W i ing—o iginal d a , and RE:
W i ing— e iew & edi ing.
Se ena Bona e i: C, FA, M, PA, S, SU, W . Moj aba Ba zega i:
C,S. Melissa Be e s: RE. S e en Boyd: S,RE. And ew J. Bu gh-
a d : RE. Donnie Came on: S, RE. F ancesco Chiumen o: RE.
Gianluigi C imi: S, RE. Ge a d Degenha : RE. Pholpa Du ong-
bhan: RE. Michelle Alejand a Espinosa He nandez: RE. Giulia
F a e igo: S, V, RE. Ali Ghasem-Zadeh: RE. Lo enzo G assi:
RE. Jukka Hi asniemi: S, RE. Mahdi Hosseini aba abaei: RE.
Gianluca Io i: DC, M, S, SU. Joe i Kok: RE. Michael Kuczynski:
S, RE. YoungJun Lee: RE. Cecilia Libe a i: RE. Sa ah Manske:
C, RE. Ma McCo mick: C, S, RE. Ma ia Monzon: S, RE. Ma -
ino Pani: RE. Simone Poncioni: S, RE. Jilmen Quin iens: RE.
Sabine R¨aube : RE. Paul Ri sche: RE. Al onso Da io San ama-
ia: S, V, RE. F ancesco San ini: C, FA, PA, S, RE. Fabio Sa o:
RE. En ico Schileo: C, SU, RE. Vincen S adelmann: FA, C, RE.
Ka h yn S. S ok: C, SU, RE. Rachel Su owiec: RE. Ful ia Taddei:
C, RE. Ja ed Vico y: S, RE. Ma hias Walle: RE. Ma iska Wesse-
ling: S. Danielle Whi ie : RE, SU. Be ina Willie: RE. Andy Kin
On Wong: SU, RE. Dˇzenan Zuki´c: S, RE.
Acknowledgmen s
We hank all he indi iduals who ha e con ibu ed o and suppo -
ed he open-sou ce ecosys em, including de elope s, main aine s,
and ad oca es, whose e o s ha e made open and ep oducible
esea ch possible. The de elopmen o he ORMIR communi y
and i s guidelines has been suppo ed by a NumFOCUS- unded
Jupy e Communi y Wo kshop and a Swiss Na ional Science Foun-
da ion g an (n. 221844). Finally, we acknowledge he use o
Cha GPT o edi o ial suppo , including g amma checking and
low enhancemen .
Open and ep oducible esea ch in MSK imaging: Why i ma e s and how o implemen i wi h he guidelines o he ORMIR communi y 7
Re e ences
1. O Panagio opoulou, SD Wilshin, EJ Ray ield, SJ She elbine,
and JR Hu chinson. Wha makes an accu a e and elia-
ble subjec -speci ic ini e elemen model? a case s udy o
an elephan emu . Jou nal o he Royal Socie y In e ace,
9(67):351–361, 2012. doi:10.1098/ si .2011.0323.
2. CA B assey, JD Ga dine , and AC Ki chene . Tes ing
hypo heses o he unc ion o he ca ni o an baculum using
ini e-elemen analysis. P oceedings o he Royal Socie y B:
Biological Sciences, 285(1887):20181473, 2018. doi:10.1098/
spb.2018.1473.
3. O Panagio opoulou, SD Wilshin, EJ Ray ield, SJ She elbine,
and JR Hu chinson. Wha makes an accu a e and elia-
ble subjec -speci ic ini e elemen model? a case s udy o
an elephan emu . Jou nal o he Royal Socie y In e ace,
11(20140854), 2014. doi:10.1098/ si .2014.0854.
4. Eu opean Resea ch Execu i e Agency. Open science. Acces-
sed: 27 Augus 2025. URL: h ps:// ea.ec.eu opa.eu/
open-science_en.
5. Na ional Academies o Sciences, Medicine, Policy, Global
A ai s, Boa d on Resea ch Da a, Di ision on Enginee ing,
Physical Sciences, Commi ee on Applied, Theo e ical S a i-
s ics, Boa d on Ma hema ical Sciences, e al. Rep oducibili y
and eplicabili y in science. Na ional Academies P ess, 2019.
doi:10.17226/25303.
6. DL Donoho, A Maleki, IU Rahman, M Shah am, and V S od-
den. Rep oducible esea ch in compu a ional ha monic analy-
sis. Compu ing in Science & Enginee ing, 11(1):8–18, 2008.
doi:10.1109/MCSE.2009.15.
7. MR Muna `o, BA Nosek, DVM Bishop, KS Bu on, CD Cham-
be s, N Pe cie du Se , U Simonsohn, E-Jan Wagenmake s,
JJ Wa e, and JPA Ioannidis. A mani es o o ep oducible
science. Na u e human beha iou , 1(1):0021, 2017. doi:h p:
//doi.o g/10.1038/s41562-016-0021.
8. GK Sand e, A Nek u enko, J Taylo , and E Ho ig. Ten simple
ules o ep oducible compu a ional esea ch. PLoS compu-
a ional biology, 9(10):e1003285, 2013. doi:h p://doi.o g/
10.1371/jou nal.pcbi.1003285.
9. A Rule, A Bi mingham, C Zuniga, I Al in as, S-C Huang,
R Knigh , N Moshi i, MH Nguyen, SB Rosen hal, F P´e ez,
and PW Rose. Ten simple ules o w i ing and sha ing
compu a ional analyses in jupy e no ebooks, 2019. doi:
10.1371/jou nal.pcbi.1007007.
10. J F ei e and F Chi iga i. P o enance and he di e en la-
o s o compu a ional ep oducibili y. IEEE Da a Enginee ing
Bulle in, 41(1):15, 2018. URL: h p://si es.compu e .o g/
debull/A18ma /p15.pd .
11. HA Piwowa , RS Day, and DB F idsma. Sha ing de ailed
esea ch da a is associa ed wi h inc eased ci a ion a e. PloS
one, 2(3):e308, 2007. doi:10.1371/jou nal.pone.0000308.
12. P Vandewalle. Code sha ing is associa ed wi h esea ch impac
in image p ocessing. Compu ing in Science & Enginee ing,
14(4):42–47, 2012. doi:10.1109/MCSE.2012.63.
13. Y Gil, CH Da id, I Demi , BT Essawy, RW Fulweile , JL Goo-
dall, L Ka ls om, H Lee, HJ Mills, J-H Oh, and SA Pie ce.
Towa d he geoscience pape o he u u e: Bes p ac ices o
documen ing and sha ing esea ch om da a o so wa e o
p o enance. Ea h and Space Science, 3(10):388–415, 2016.
doi:h p://dx.doi.o g/10.1002/2015EA000136.
14. V S odden. The scien i ic me hod in p ac ice: Rep oducibi-
li y in he compu a ional sciences. MIT Sloan esea ch pape ,
(4773-10), 2010. doi:10.2139/ss n.1550193.
15. J F ei e. A p o enance-based in as uc u e o c ea ing ep o-
ducible pape s. In Rep oducible Resea ch: Tools and S a egies
o Scien i ic Compu ing. A wo kshop in associa ion wi h
Applied Ma hema ics Pe spec i es 2011, Uni e si y o B i-
ish Columbia, Canada. July 13–16, 2011. URL: h ps:
//s odden.ne /AMP2011/index.h ml.
16. X Chen, S Dallmeie -Tiessen, R Dasle , S Fege , P Fokianos,
JB Gonzalez, H Hi onsalo, D Kousidis, A La asa, S Mele,
and DR Rod iguez. Open is no enough. Na u e Physics,
15(2):113–119, 2019. doi:10.1038/s41567-018-0342-2.
17. Y Pe ez-Ri e ol, L Ga o, R Wang, T Sachsenbe g, J Usz-
ko ei , Fd V Lep e os , C Fu ezan, T Te nen , SJ Eglen,
DS Ka z, and TJ Polla d. Ten simple ules o aking
ad an age o gi and gi hub. PLoS compu a ional biology,
12(7):e1004947, 2016. doi:10.1371/jou nal.pcbi.1004947.
18. A Mo in, J U ban, and P Sliz. A quick guide o so wa e
licensing o he scien is -p og amme . PLOS Compu a io-
nal Biology, 8(7):1–7, 07 2012. doi:10.1371/jou nal.pcbi.
1002598.
19. A P li´c and JB P oc e . Ten simple ules o he open de e-
lopmen o scien i ic so wa e. PLoS compu a ional biology,
8(12):e1002802, 2012. doi:10.1371/jou nal.pcbi.1002802.
20. RA Pold ack, CJ Ma kiewicz, S Appelho , YK Asha , T Aue ,
S Baille , S Bansal, L Bel achini, C G Bena , G Be azzoli,
e al. The pas , p esen , and u u e o he b ain imaging
da a s uc u e (bids). Imaging Neu oscience, 2:1–19, 2024.
doi:10.1162/imag_a_00103.
21. RC Gen leman, VJ Ca ey, DM Ba es, B Bols ad, M De -
ling, S Dudoi , B Ellis, L Gau ie , Y Ge, J Gen y, e al.
Bioconduc o : open so wa e de elopmen o compu a ional
biology and bioin o ma ics. Genome biology, 5(10):R80, 2004.
doi:10.1186/gb-2004-5-10- 80.
22. TE Odaka, A Banihi we, G Eyna d-Bon emps, A Pon e,
G Maze, K Paul, J Bake , and R Abe na hey. The Pan-
geo ecosys em: in e ac i e compu ing ools o he geosciences:
benchma king on HPC. In Annual Wo kshop on HPC Use
Suppo Tools, pages 190–204. Sp inge , 2019. doi:10.1007/
978-3-030-44728-1_12.
23. S Bona e i, L Ba zaghi, MM Ba zega i, AJ Bu gha d ,
D Came on, J Ca ballido-Gamio, G C imi, P Du ongbhan,
M Espinosa He nandez, G F a e igo, L G assi, H G ee ,
G Io i, M Kuczynski, S Manske, M McCo mick, N Nee eson,
M Nie hamme , M Pani, J Quin iens, MM Sadeghi, F San ini,
E Schileo, KS S ok, F Taddei, JJ Tse, J Vico y, M Wesseling,
AKO Wong, and D Zuki´c. In oducing he open and ep odu-
cible musculoskele al imaging esea ch (ORMIR) communi y,
July 2023. doi:10.5281/zenodo.8119243.
24. G Wilson, J B yan, K C ans on, J Ki zes, L Nede b ag , and
TK Teal. Good enough p ac ices in scien i ic compu ing. PLoS
compu a ional biology, 13(6):e1005510, 2017. doi:10.1371/
jou nal.pcbi.1005510.
25. S Bona e i, MA Espinosa He nandez, F Chiumen o, Y Fou-
nas, M F oeling, J Hi asniemi, G Io i, Y Lee, S Ma uschik,
M Monzon, F San ini, and D Came on. ORMIR-MIDS: An
open s anda d o cu a ing and sha ing musculoskele al ima-
ging da a. In 24 h In e na ional Wo kshop on Quan i a i e
Musculoskele al Imaging (QMSKI), The Ba ossa Valley, Sou h
Aus alia, No embe 2024. doi:10.1093/jbm pl/zia 035.
8Se ena Bona e i e al.
26. K zysz o J Go golewski, Tibo Aue , Vince D Calhoun,
R Came on C addock, Sami Das, Eugene P Du , Guillaume
Flandin, Sa aji S Ghosh, T is an Gla a d, Ya osla O Halch-
enko, e al. The b ain imaging da a s uc u e, a o ma
o o ganizing and desc ibing ou pu s o neu oimaging expe-
imen s. Scien i ic da a, 3(1):1–9, 2016. doi:10.1038/sda a.
2016.44.
27. WD Bidgood J and SC Ho ii. In oduc ion o he ACR-
NEMA DICOM s anda d. Radiog aphics, 12(2):345–355, 1992.
doi:10.1148/ adiog aphics.12.2.1561424.
28. RW Cox, J Ashbu ne , H B eman, K Fissell, C Haselg o e,
CJ Holmes, JL Lancas e , DE Rex, SM Smi h, JB Woodwa d,
e al. A (so o ) new image da a o ma s anda d: NiFTI-1.
In 10 h annual mee ing o he o ganiza ion o human b ain
mapping, olume 22, page 01, 2004.
29. KJ Millman and F P´e ez. De eloping open-sou ce scien i ic
p ac ice. In Implemen ing ep oducible esea ch, pages 149–
183. Chapman and Hall/CRC, 2018. URL: h ps://doi.o g/
10.1201/9781315373461.
30. M McCo mick, X Liu, J Jomie , C Ma ion, and L Ibanez. I k:
enabling ep oducible esea ch and open science. F on ie s
in neu oin o ma ics, 8:13, 2014. doi:10.3389/ nin .2014.
00013.
31. BC Lowekamp, DT Chen, L Ib´a˜nez, and D Blezek. The design
o simplei k. F on ie s in neu oin o ma ics, 7:45, 2013. doi:
10.3389/ nin .2013.00045.
32. A Paszke, S G oss, F Massa, A Le e , J B ad-
bu y, G Chanan, T Killeen, Z Lin, N Gimelsh-
ein, L An iga, and A Desmaison. Py o ch: An
impe a i e s yle, high-pe o mance deep lea ning lib a y.
Ad ances in neu al in o ma ion p ocessing sys ems, 32,
2019. URL: h ps://p oceedings.neu ips.cc/pape /2019/
hash/bdbca288 ee7 92 2b a9 7012727740-Abs ac .h ml.
33. M Abadi, A Aga wal, P Ba ham, E B e do, Z Chen,
C Ci o, GS Co ado, A Da is, J Dean, M De in, e al.
Tenso low: La ge-scale machine lea ning on he e ogeneous
sys ems, 2015. URL: h ps://www.usenix.o g/sys em/
iles/con e ence/osdi16/osdi16-abadi.pd .
34. F Ped egosa, G Va oquaux, A G am o , V Michel, B Thi ion,
O G isel, M Blondel, P P e enho e , R Weiss, V Dubou g,
J Vande plas, A Passos, D Cou napeau, M B uche , M Pe o ,
and E Duchesnay. Sciki -lea n: Machine lea ning in Py hon.
Jou nal o Machine Lea ning Resea ch, 12:2825–2830, 2011.
35. Py hon Packaging Au ho i y. Py hon packaging use guide.
Accessed: 28 Augus 2025. URL: h ps://packaging.py hon.
o g/en/la es /.
36. DE Knu h. Li e a e p og amming. The compu e jou nal,
27(2):97–111, 1984. doi:10.1093/comjnl/27.2.97.
37. G an Rossum, B Wa saw, and A Coghlan. Pep8 - S yle
guide o Py hon code. Accessed: 28 Augus 2025. URL:
h ps://peps.py hon.o g/pep-0008/.
38. Numpy s yle guide. Accessed: 28 Augus 2025. URL: h ps:
//numpydoc. ead hedocs.io/en/la es / o ma .h ml.
39. Google Py hon s yle guide. Accessed: 28 Augus 2025. URL:
h ps://google.gi hub.io/s yleguide/pyguide.h ml.
40. FdV Lep e os , VC Ba bosa, EL F ancisco, Y Pe ez-Ri e ol,
and PC Ca alho. On bes p ac ices in he de elopmen o
bioin o ma ics so wa e. F on ie s in gene ics, 5:199, 2014.
doi:10.3389/ gene.2014.00199.
41. A Tu ne , B T an, C Sewell, F F ei ag, JL Ande sen, JF B.,
S Finucane, T Shimizukawa, and T Komiya. Sphynx. Acces-
sed: 28 Augus 2025. URL: h ps://www.sphinx-doc.o g/en/
mas e /index.h ml.
42. Read he Docs. Accessed: 28 Augus 2025. URL: h ps:
//abou . ead hedocs.com/.
43. Execu able Books Communi y. Jupy e Book. Accessed: 28
Augus 2025. URL: h ps://jupy e book.o g/en/s able/
in o.h ml.
44. J Scheibne , M Ienca, S Kechagia, JR T oncoso-Pas o iza,
JL Raisa o, J-P Hubaux, J Fellay, and E Vayena. Da a
p o ec ion and e hics equi emen s o mul isi e esea ch wi h
heal h da a: a compa a i e examina ion o legisla i e go e na-
nce amewo ks and he ole o da a p o ec ion echnologies

.
Jou nal o Law and he Biosciences, 7(1):lsaa010, 05 2020.
doi:10.1093/jlb/lsaa010.
45. JC Ba lle, K D eye , B Allen, T Cook, CJ Ro h, AB Ki s,
R Geis, CC Wu, MP Lung en, J Pa i, e al. Da a sha-
ing o imaging in an e ol ing heal h ca e wo ld: epo o
he ACR da a sha ing wo kg oup, pa 1: da a e hics o
p i acy, consen , and anonymiza ion. Jou nal o he Ame-
ican College o Radiology, 18(12):1646–1654, 2021. doi:
10.1016/j.jac .2021.07.014.
46. A Bu gha d , AKO Wong, F Taddei, K S ok, S Manske,
S Bona e i, and V S adelmann. ORMIR da a sha ing gui-
delines - 1.1 ( 1.1). Zenodo, 2025. doi:10.5281/zenodo.
17206129.
47. P Ri sche, F Sa o, F San ini, C Lei ne , M F anchi, O Faude,
and N C onin. UMUD: A web applica ion o easy access
o musculoskele al ul asonog aphy da ase s, Janua y 2025.
P ep in . doi:10.31219/os .io/sy 4z.
48. RP Della alle, EJ Hes e , LF Heilig, AL D ake, JW Kun z-
man, M G abe , and LM Schilling. Going, going, gone:
Los in e ne e e ences. Science, 302(5646):787–788, 2003.
doi:10.1126/science.1088234.
49. M Klein, H Van de Sompel, R Sande son, H Shanka , L Balaki-
e a, K Zhou, and R Tobin. Schola ly con ex no ound: one in
i e a icles su e s om e e ence o . PloS one, 9(12):e115253,
2014. doi:10.1371/jou nal.pone.0115253.
50. Jupy e P ojec , M Bussonnie , J Fo de, J F eeman, B G an-
ge , T Head, C Holdg a , K Kelley, G Nal a e, A Oshe o ,
M Pace , Y Panda, F P´e ez, B Ragan-Kelley, and C Willing.
Binde 2.0— ep oducible, in e ac i e, sha able en i onmen s
o science a scale. Scipy, 2018. URL: h ps://p oceedings.
scipy.o g/a icles/Majo a-4a 1 417-011.pd .
51. Gi Hub Pages documen a ion. Accessed: 28 Augus 2025.
URL: h ps://docs.gi hub.com/en/pages.
52. Jupy e Book - Publish you book online. Accessed: 28
Augus 2025. URL: h ps://jupy e book.o g/en/s able/
s a /publish.h ml.
53. Gi Hub Docs - Re e encing and ci ing con en . Acces-
sed: 28 Augus 2025. URL: h ps://docs.gi hub.
com/en/ eposi o ies/a chi ing-a-gi hub- eposi o y/
e e encing-and-ci ing-con en .
54. How o connec Figsha e wi h you Gi Hub
accoun . Accessed: 28 Augus 2025.
URL: h ps://in o. igsha e.com/use -guide/
how- o-connec - igsha e-wi h-you -gi hub-accoun /.
55. Welcome o so wa e he i age documen a ion - ge a pe -
manen iden i ie o you sou ce code. Accessed: 28
Augus 2025. URL: h ps://docs.so wa ehe i age.o g/
Open and ep oducible esea ch in MSK imaging: Why i ma e s and how o implemen i wi h he guidelines o he ORMIR communi y 9
#landing-basics.
56. JF Clae bou and M Ka enbach. Elec onic documen s gi e
ep oducible esea ch a new meaning. In SEG echnical p o-
g am expanded abs ac s 1992, pages 601–604. Socie y o
Explo a ion Geophysicis s, 1992. doi:10.1190/1.1822162.
57. G Cap a elli, B Sedo a, M Ricci, S S all, and M Giampoala.
No ebooks now! he u u e o ep oducible esea ch. Ea h and
Space Science, 10(12):e2023EA003458, 2023. doi:10.1029/
2023EA003458.
58. C E dmann, S S all, C Gen emann, C Holdg a , FPA Fe -
nandes, Gehlen KP- on, and M Co ellec. Guidance o agu
au ho s - jupy e no ebooks, May 2021. doi:10.5281/zenodo.
5651648.