An interactive and intuitive visualisation method for X-ray computed tomography data of biological samples in 3D Portable Document Format

1 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s An in e ac i e and in ui i e isualisa ion me hod o X- ay compu ed omog aphy da a o biological samples in 3D Po able Documen Fo ma Ma ké a Tesařo á 1, eglan ine Heude 2,3,4, Glenda comai 3,4, Tomáš Zikmund 1, Ma ké a Kaucká 5,6, igo Adameyko5,6, Shah agim ajbakhsh 3,4 & Joze Kaise 1* 3D imaging app oaches based on X- ay mic ocompu ed omog aphy (mic oCT) ha e become inc easingly accessible wi h ad ancemen s in me hods, ins umen s and expe ise. he syne gy o ma e ial and li e sciences has impac ed biomedical esea ch by p oposing new ools o in es iga ion. Howe e , da a sha ing emains challenging as mic oCT iles a e usually in he ange o gigaby es and equi e speci ic and expensi e so wa e o ende ing and in e p e a ion. He e, we p o ide an ad anced me hod o isualisa ion and in e p e a ion o mic oCT da a wi h small ile o ma s, eadable on all ope a ing sys ems, using eely a ailable Po able Documen Fo ma (PDF) so wa e. Ou me hod is based on he con e sion o olume ic da a in o in e ac i e 3D PDF, allowing o a ion, mo emen , magni ica ion and se ing modi ica ions o objec s, hus p o iding an in ui i e app oach o analyse s uc u es in a 3D con ex . We desc ibe he comple e pipeline om da a acquisi ion, da a p ocessing and comp ession, o 3D PDF o ma ing on an example o c anio acial ana omical mo phology in he mouse emb yo. ou p ocedu e is widely applicable in biological esea ch and can be used as a amewo k o analyse olume ic da a om any esea ch ield elying on 3D ende ing and CT- biomedical imaging. One o he o midable phenomena in de elopmen al biology is how he shape di e si y obse ed among li ing o ganisms is de ined and con olled du ing de elopmen and g ow h. Emb yonic pa e ning is a highly dynamic p ocess implica ing mul iple molecula mechanisms and cell in e ac ions a he basis o o gan o ma ion. In he human emb yo, de ec s in such cellula p ocesses can a ec he de elopmen al p og am leading o congeni al diso de s. Congeni al de ec s ha e an incidence o 3% in he human popula ion1 and hey a e causal o up o one-qua e o all epo ed neona al dea hs2. Thus, con ex ual isualisa ion o emb yonic de elopmen is c i ical o elucida e he o igins o mal o ma ions. Mul i-disciplina y collec i es composed o clinical doc o s, biologis s, enginee s and imaging expe s a e cu en ly pushing o wa d he unde s anding o biological ques ions using h ee dimensional (3D) app oaches. While he analysis o his ological sec ions emains a mains ay in de elopmen al biology, he econs uc ion o 3D olumes om 2D slices has p o ided impo an in o ma ion o unde s and mo phogenesis in mouse and human emb yos3–6. Non-des uc i e echnologies such as 3D imaging by con ocal mic oscopy and ligh shee me hods7–9, op ical p ojec ion omog aphy (OPT)10–12 and mic o-compu ed omog aphy (mic oCT)13–16 cons i- u e eme ging powe ul me hods o analyse he opog aphy o de eloping s uc u es in 3D con ex , o gain insigh in o he pa hogenesis o congeni al diso de s. Howe e , one majo challenge o 3D app oaches is sha ing o com- plex da ase s e ec i ely and in ui i ely be ween colleagues om di e en ields o discussion, and ul ima ely 1Cen al Eu opean Ins i u e o Technology, B no Uni e si y o Technology, B no, Czech Republic. 2Depa men Adap a ion du Vi an , Museum na ional d’His oi e na u elle, CNRS UMR 7221, Pa is, F ance. 3Depa men o De elopmen al and S em Cell Biology, S em Cells and De elopmen Uni , Ins i u Pas eu , Pa is, F ance. 4CNRS UMR, 3738, Pa is, F ance. 5Depa men o Physiology and Pha macology, Ka olinska Ins i u e , Solna, Sweden. 6Depa men o Molecula Neu osciences, Medical Uni e si y o Vienna, Vienna, Aus ia. *email: kaise @ me. u b .cz open 2 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ p esen ing hem in a publica ion o ma . Gene ally, 3D da ase s need special so wa e o isualisa ion o a e educed o 2D pic u es in which impo an in o ma ion migh be los . The s udy o he de eloping head cons i u es a good example o he need o 3D app oaches gi en i s com- plex ana omy. The c anio acial egion is buil o di e se emb yonic cell ypes gi ing ise o ha d and so issues including skele al, muscula and ne ous componen s17. The inal shape o he ace s ongly depends on he geome y o heskele al elemen s and hei in e ac ion wi h adjacen so issues such as muscles and he ne ous sys em15,16. Nume ous congeni al c anio acial abno mali ies a ec he o m and unc ion o he ace and explana- ions o  hese mal o ma ions s ill awai he undamen al unde s anding o he unde lying ailu e o mo phogene- sis18. The mic oCT app oach ollowed by 3D econs uc ion has enabled high-quali y in o ma ion o he complex mo phological aspec s o head and ace de elopmen 15,16. To acili a e assimila ion and isualisa ion o mic oCT da ase s, in e ac i e 3D Po able Documen Fo ma (PDF) has been used in di e en ield including in de elopmen al biology6,19,20, in human physiology and ana - omy4,21–23, in en omology24 and ma ine biology25,26. Howe e , a majo limi a ion o p e iously published me hods is he equi emen o ad anced p og amming skills and/o ins alla ion o u he p epaid so wa e packages21. Mos o hese app oaches depend on he use o Adobe Ac oba P o Ex ended so wa e4,6,19,23–28 o Ja aSc ip p og amming language20,27. He e, we p o ide an al e na i e, use - iendly way o c ea e in e ac i e 3D PDF iles om mic oCT da ase s aking he complexi y o mouse c anio acial ana omy as a model example. Ou inno a i e and e icien pipeline comp ises mic oCT da a acquisi ion, segmen a ion and inal es ablishmen o a 3D PDF using a combina ion o ee and p e-paid so wa e. The esul ing ile can be iewed wi h s anda d PDF iewe s and o e s an in e ac i e in e ace o mic oCT da a sha ing, analysis and p esen a ion. Pipeline o he C ea ion o In e ac i e 3D PDF To c ea e an in e ac i e 3D PDF om mic oCT da a, he c i ical s eps a e: (i) chemical con as ing o biological samples i so issues a e o be isualized, (ii) da a acquisi ion and (iii) CT i ual econs uc ion (Fig.1). This is ollowed by da a p ocessing, he mos c i ical s ep consis ing o segmen a ion, su ace ex ac ion, adjus men o segmen ed 3D models, and con e sion in o a inal in e ac i e PDF ile. He e, we p o ide a de ailed desc ip ion o each s ep o he pipeline (see also Supplemen a y Ma e ial1). Figu e 1. O e iew o he me hod pipeline desc ibed in his s udy. 3 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ X-Ray Mic o omog aphy Measu emen Mic oCT is an es ablished echnology o imaging mine alized issues in animal specimens. Howe e , i s use in compa a i e mo phology has been limi ed by he low in insic X- ay con as o non-mine alised so issues. To o e come his p oblem, me hods ha e been de eloped o inc ease issue con as , including chemical ea men wi h con as ing agen s29,30, phase-con as imaging31–36 o dual-ene gy compu ed omog aphy (DECT)37. Chemical con as ing ea men s ha e mos ly been applied o he cha ac e iza ion o musculoskele al issues on small ixed samples. Fo in i o CT imaging including clinical diagnoses, non- oxic iodine-based con as agen s (e.g. iohexol o hexab ix) a e used o he analysis o he ca dio ascula sys em and ca i ies, bu no o di ec analysis o muscle issues38,39. Clinical CT imaging o he musculoskele al sys em is commonly pe o med wi hou a con as ing agen , he eby limi ing he analysis o disc e e so o gans40. The e o e, on small ixed sam- ples, mul iple con as ing p o ocols ha e been used, each wi h i s own ad an ages and limi a ions41,42. The s ain- ings based on iodine, osmium o he oxic phospho ungs ic acid (PTA) a e he mos commonly used29,30. Fo he pipeline p oposed he e on he s udy o c anio acial mouse de elopmen , we p opose he use o PTA con as ing ea men ha pe mi s high con as imaging o a wide a ie y o so and mine alised issues composing he mouse head (Fig.2). Howe e , mic oCT da a ob ained wi h o he con as ing agen s o in hei absence, such as clinical da a o da a om non-biological specimens, will be equally ameneable o he subsequen s eps o he 3DPDF pipeline. Da a P ocessing Following he acquisi ion o high-con as omog aphic slices, he segmen a ion o complex c anio acial s uc- u es cons i u es a majo challenge. Fully-au oma ic app oaches ha e been ied and we e gene ally unsuccess ul43. Thus, he manual app oach is usually he only me hod a ailable o achie e p ecise and accu a e segmen a ions44. Howe e , some semi-au oma ic me hods such as local segmen a ion o in e pola ion be ween manually seg- men ed slices can be used45,46. Fo e i ica ion o he accu acy o manually segmen ed s uc u es, Fig.2 shows bo h o iginal and segmen ed omog aphic slices. This app oach pe mi ed he segmen a ion o so issues such as eyeballs, ex aocula muscles and he cen al ne ous sys em, as well as o ha d issues including he chond o- c anium, bones and u u e ee h (den al placodes) (Fig.3). Fo 3D olume ende ing, i is hen mo e con enien o use 3D mesh o ma s a he han a s ack o 2D images. 3D mesh o ma s ep esen a se ies o 2D polygons ( ypically iangles o quad angles) linked oge he o ec e- a e he su ace o a 3D objec 28. I encodes he 3D model’s geome y, colou s, ex u es, e c. Fo u he wo k, he segmen ed masks need o be ans e ed o he meshes. Mos so wa e can ans e he mask in o a wide ange o Figu e 2. Tomog aphic measu emen s and segmen a ion o c anio acial s uc u es in a mouse emb yo 15 days pos - e ilisa ion. Colou planes on 3D models (le panels) indica e he posi ion o he aw omog aphic slices (middle panels) and some segmen ed s uc u es ( igh panels) including he cen al ne ous sys em (pu ple), he lens (da k blue), he den al placodes (ligh blue) and ex aocula muscles ( ed). 4 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ equi alen o ma s. Howe e , only some o he o ma s con ain in o ma ion abou he colou (e.g. OBJ, VRML and STEP) which is impo an o he isualisa ion o se e al s uc u es in one model. O he o ma s ep esen only he mesh (e.g. STL and Ma lab m- ile). Ano he p oblem o be sol ed is he size o he model (i.e. he numbe o aces in he mesh). When ans e - ing he segmen ed mask in o he mesh, an unnecessa ily la ge numbe o aces can be c ea ed. Compa ison o a di e en numbe o aces and co esponding size can be ound in Fig.4. Su p isingly, educing by ou imes he numbe o aces does no signi ican ly comp omise de ails in he 3D models. Simple shapes such as he eyeball do no show no able e o s besides he sligh de o ma ion o he sphe e. Howe e , u he simpli ica ion o complex shapes, such as he chond oc anium and ex aocula muscles (EOM), educes he model quali y and some de ails a e los (Fig.4, ed a ows). Thus, depending on he s uc u es analysed and he esolu ion needed, di e en ace educ ions should be es ed and alida ed be o e u he analysis. The e o e, he comp ession o he da a should be cus omised acco ding o he complexi y o he model and he esolu ion o omog aphic slices. Among he 3D p in ing so wa e, some ee so wa e enable colou modi ica ions o mesh simpli ica ion. He e, we made use o he Meshlab47 and Blende 48 packages o his pu pose. No ably, he la e allows uni ying di e en sub-models in o one mesh p ese ing he indi iduali y o each objec (see Supplemen a y Ma e ial1 o mo e de ails). C ea ion o an In e ac i e 3D PDF Once he models a e colou -labelled and simpli ied, hey a e hen con e ed in o an in e ac i e ile. We exploi ed 3D PDF Make S andalone49 aking ad an age o he p e-p epa ed model ha p o ides an in e ac i e window in he PDF ile by clicking on Add 3D bu on. Be o e embedding he model in o an in e ac i e ile, he use can p epa e an o dina y PDF ile using s anda d so wa e (e.g. Mic oso O ice, Apache OpenO ice e c.). This ile is used as a empla e showing, o example, indi idual s uc u es o p ede ined iews. These a eas (i.e. images o signs) a e aken in o li e in 3D PDF Make by assigning wha e e a ea in he page o speci ic iew/s uc u e (see o mo e de ails Supplemen a y Ma e ials1 and 2). The model documen p esen ed he e (Supplemen a y Ma e ial3) consis s o ou p ede ined iews ( en al, do sal, la e al and medial) and a possibili y o showing indi idual c anio acial s uc u es (b ain, ca ilage, bone and ee h). I helps o manipula e he 3D model and can be se acco ding o he a eas/ iews o in e es ha a e impo an o show. Fo example, o be e isualisa ion, some s uc u es can be se o he semi- anspa en mode, o s uc u es can be swi ched on/o indi idually in he model ee. The in e ac i e 3D PDF also allows he use o o a e, u n and pan he model, change he p e-de ined 3D- ende ing, change he ligh s and backg ound colou o add a i ual c oss-sec ion on he model. Figu e 3. Su ace ende ing o segmen ed s uc u es in a mouse emb yo 15 days pos - e ilisa ion. S uc u es o in e es a e colou -coded. 5 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ Discussion and Conclusion Compa a i e mo phological s udies in he ield o de elopmen al biology ha e been challenging, bu con as -enhanced X- ay compu ed omog aphy has b ough new possibili ies o high- esolu ion 3D isualis- a ion15,16,50. In his s udy, we p esen a de ailed, use - iendly p o ocol o he su ace ende ing o c anio acial s uc u es including so issues (muscles, eyeballs, cen al ne ous sys em) as well as o ha d issues (ca ilages, bones and ee h), wi h a s ep-by-s ep p ocedu e om sample collec ion o he c ea ion o an in e ac i e 3D PDF. The inal PDF is eadable on all ope a ing sys ems wi h he ee s anda d AdobeReade ®/Ac oba Reade DC (Adobe Inc., Cali o nia USA). Ou 3D econs uc ion me hodology has al eady shown i s u ili y and s eng h o isualisa ion and pheno- ypic analysis o complex s uc u es such as he neck musculoskele al sys em50 and nasal capsules o con ol and mu an mouse emb yos16. The e o e, a use - iendly me hod o c ea ing such iles will be o g ea u ili y o biol- ogis s. In addi ion, ou p ocedu e can be used as a amewo k o analyse olume ic da a o any ield o esea ch ha elies on 3D ende ing, e.g. o isualisa ion o olume ic in o ma ion o geological samples by lase -induced b eakdown spec oscopy51. The use o in e ac i e 3D PDF iles has a g ea po en ial o da a sha ing, communica ion and publica ions21, bu is s ill spa sely used. We belie e ha ou wo k will inspi e esea che s wo king wi h 3D imaging o p esen hei da a in such a use - iendly o ma . Me hods Use o expe imen al animals. All animal wo k was app o ed and pe mi ed by he Local E hical Commi ee on Animal Expe imen s (No h S ockholm Animal E hics Commi ee) and conduc ed acco ding o The Swedish Animal Agency’s P o isions and Guidelines o Animal Expe imen a ion ecommenda ions. Mice we e sac i iced wi h iso lu ane (Bax e KDG9623) o e dose o ce ical disloca ion, and emb yos we e collec ed in o ice-cold PBS. Subsequen ly, issue samples we e ixed in o eshly p epa ed 4% pa a o maldehyde (PFA) in PBS solu ion o 24 hou s a +4 °C wi h slow o a ion and washed in PBS. Figu e 4. The numbe o aces a ec s he de ec ion o de ails in chond oc anium, ex aocula muscles (EOM) and eyeball models. Red a ows indica e de ails in he 3D model ha disappea wi h model simpli ica ion. 6 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ issue con as ing. S aining p o ocol has been adap ed and modi ied om he o iginal p o ocol de eloped by B ian Me sche labo a o y33,34. A e ixa ion, he samples we e dehyd a ed in inc easing concen a ion o e h- anol se ies (30%, 50%, 70%, 90%), one day in each concen a ion o minimise he sh inkage o issues. Fo issue con as ing, samples we e hen ans e ed in o 1% PTA (phospho ungs ic acid) in 90% me hanol o h ee weeks. The PTA-me hanol solu ion was changed e e y wo days. Subsequen ly, he samples we e ehyd a ed by e hanol se ies (90%, 70%, 50% and 30%) and shipped o he CT-labo a o y o scanning. X- ay mic oc measu emen s. A e ixa ion and con as ing ea men s, samples we e ully ehyd a ed in dis illed wa e , embedded in 1.0% aga ose gel and placed in a polyp opylene ube o a oid mo emen a e ac s du ing omog aphy scanning. The polyp opylene ube was ixed on a plas ic od by a silicone gun. The od was pu in he cen e o he o a ion s age axis. The mic oCT scanning was pe o med using he labo a o y sys- em GE Phoenix | ome|x L 240 (GE Sensing & Inspec ion Technologies GmbH, Ge many) wi h a 180 kV/15 W maximum powe nano ocus X- ay ube and la panel dynamic 41|100 wi h 4000 × 4000 px and a pixel size o 100 × 100 μm. The exposu e ime was 600 ms in each o he 2200 p ojec ions acqui ed o e a o al angle o 360°. Th ee p ojec ions we e acqui ed and a e aged in e e y posi ion o educe he noise in he omog aphic da a. Thus, he scanning ime was 73 minu es. The accele a ion ol age and cu en o he X- ay ube we e 60 kV and 200 μA, espec i ely. The beam was il e ed by a 0.2 mm- hick aluminium il e . The linea oxel esolu ion o he measu emen was se o 4.2 μm in all dimensions. The omog aphic econs uc ion was conduc ed using he so - wa e GE phoenix da os|x 2.0 (GE Sensing & Inspec ion Technologies GmbH, Ge many). Da a p ocessing and analysis. Segmen a ion o c anio acial s uc u es was pe o med semi-manually in A izo (The mo Fishe Scien i ic, USA). We used some au oma ic ools using a egion g owing me hod and h esholding on 2D slices. Segmen a ions we e done on one o h ee o i e slices, and he mask was in e pola ed. Subsequen ly, he segmen ed models we e smoo hed in VG S udio MAX 3.2 (Volume G aphics GmbH, Ge many) acco ding o45. The segmen ed mask was hen expo ed in o a mesh (*.OBJ o ma ) and was colou -labelled and adjus ed in Meshlab47 and Blende so wa e48. The colou -coded models we e ans e ed in o a p e-p epa ed PDF ile in 3D PDF Make S andalone49. The de ailed manual can be ound in Supplemen a y Ma e ial1. The use o he in e ac i e 3D PDF is a ailable in Supplemen a y Ma e ial2. Da a a ailabili y The da ase s gene a ed du ing and/o analysed du ing he cu en s udy a e a ailable om he co esponding au ho on easonable eques . Recei ed: 20 May 2019; Accep ed: 25 Sep embe 2019; Published: xx xx xxxx Re e ences 1. Wo ld Heal h O ganiza ion. Congeni al anomalies, Fac shee No. 370. h ps://www.who.in /news- oom/ ac -shee s/de ail/ congeni al-anomalies (2016). 2. Wo ld Heal h O ganiza ion. Bi h de ec s: epo by he Sec e a ia , Execu i e boa d 125 h session EB125/7. h p://www.who.in / i is/handle/10665/2271 (2009). 3. Weninge , W. e al. High- esolu ion episcopic mic oscopy: a apid echnique o high de ailed 3D analysis o gene ac i i y in he con ex o issue a chi ec u e and mo phology. Ana Emb yol 211, 213–221, h ps://doi.o g/10.1007/s00429-005-0073-x (2006). 4. De Bakke , B. e al. An in e ac i e h ee-dimensional digi al a las and quan i a i e da abase o human de elopmen . Science 354, 1019–1028, h ps://doi.o g/10.1126/science.aag0053 (2016). 5. De Bakke , B., De Jong, K., Hagoo , J., Oos a, R. & Moo man, A. Towa ds a 3-dimensional a las o he de eloping human emb yo: The Ams e dam expe ience. Rep oduc i e Toxicol 34, 225–236, h ps://doi.o g/10.1016/j. ep o ox.2012.05.087 (2012). 6. De Boe , B., Van den Be g, G., De Boe , P., Moo man, A. & Ruij e , J. G ow h o he de eloping mouse hea : An in e ac i e quali a i e and quan i a i e 3D a las. De Biol 368, 203–213, h ps://doi.o g/10.1016/j.ydbio.2012.05.001 (2012). 7. Belle, M. e al. T idimensional Visualiza ion and Analysis o Ea ly Human De elopmen . Cell 169, 161–173, h ps://doi. o g/10.1016/j.cell.2017.03.008 (2017). 8. Renie , N. e al. iDISCO: A Simple, Rapid Me hod o Immunolabel La ge Tissue Samples o Volume Imaging. Cell 159, 896–910, h ps://doi.o g/10.1016/j.cell.2014.10.010 (2014). 9. Ragazzi e al. G. Fluo escence con ocal mic oscopy o pa hologis s. Mod Pa hol 27, 460–471, h ps://doi.o g/10.1038/ modpa hol.2013.158 (2014). 10. Dickinson, M. e al. High- h oughpu disco e y o no el de elopmen al pheno ypes. Na u e 537, 508–514, h ps://doi.o g/10.1038/ na u e19356 (2016). 11. Sha pe, J. Op ical P ojec ion Tomog aphy as a Tool o 3D Mic oscopy and Gene Exp ession S udies. Science 296, 541–545, h ps:// doi.o g/10.1126/science.1068206 (2002). 12. Sha pe, J. Op ical p ojec ion omog aphy as a new ool o s udying emb yo ana omy. J Ana 202, 175–181, h ps://doi.o g/10.1046/ j.1469-7580.2003.00155.x (2003). 13. Wong, M., Do , A., Walls, J., Le ch, J. & Henkelman, R. A no el 3D mouse emb yo a las based on mic o-CT. De 139, 3248–3256, h ps://doi.o g/10.1242/de .082016 (2012). 14. Hsu, C. e al. Th ee-dimensional mic oCT imaging o mouse de elopmen om ea ly pos -implan a ion o ea ly pos na al s ages. De Biol 419, 229–236, h ps://doi.o g/10.1016/j.ydbio.2016.09.011 (2016). 15. Kaucka, M. e al. O ien ed clonal cell dynamics enables accu a e g ow h and shaping o e eb a e ca ilage. eli e 6, e25902, h ps:// doi.o g/10.7554/eLi e.25902 (2017). 16. Kaucka, M. e al. Signals om he b ain and ol ac o y epi helium con ol shaping o he mammalian nasal capsule ca ilage. eLi e 7, e34465, h ps://doi.o g/10.7554/eLi e.34465 (2018). 17. Noden, D. & T aino , P. Rela ions and in e ac ions be ween c anial mesode m and neu al c es popula ions. J Ana 207, 575–601, h ps://doi.o g/10.1111/j.1469-7580.2005.00473.x (2005). 18. Buchanan, E., Xue, A. & Hollie , L. C anio acial Synd omes. Plas Recons Su g 134, 128e–153e, h ps://doi.o g/10.1097/ PRS.0000000000000308 (2014). 7 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ 19. De Boe , B. e al. The in e ac i e p esen a ion o 3D in o ma ion ob ained om econs uc ed da ase s and 3D placemen o single his ological sec ions wi h he 3D po able documen o ma . De 138, 159–167, h ps://doi.o g/10.1242/de .051086 (2010). 20. De Lau ie , A. e al. The Mouse Limb Ana omy A las: An in e ac i e 3D ool o s udying emb yonic limb pa e ning. BMC De Biol 8, 1–7, h ps://doi.o g/10.1186/1471-213x-8-83 (2008). 21. Newe, A. & Becke , L. Th ee-Dimensional Po able Documen Fo ma (3D PDF) in Clinical Communica ion and Biomedical Sciences: Sys ema ic Re iew o Applica ions, Tools, and P o ocols. JMIR Med In o m 6, e10295, h ps://doi.o g/10.2196/10295 (2018). 22. Danz, J. & Ka sa os, C. Th ee-dimensional po able documen o ma : A simple way o p esen 3-dimensional da a in an elec onic publica ion. Am J O hod Den o ac O hop 140, 274–276, h ps://doi.o g/10.1016/j.ajodo.2011.04.010 (2011). 23. Vale a-Melé, M. e al. A No el and F eely A ailable In e ac i e 3d Model o he In e nal Ca o id A e y. J Med Sys 42, 6, h ps://doi. o g/10.1007/s10916-018-0919-4 (2018). 24. Van de Kamp, T. e al. Th ee-Dimensional Recons uc ions Come o Li e – In e ac i e 3D PDF Anima ions in Func ional Mo phology. PLoS ONE 9, e102355, h ps://doi.o g/10.1371/jou nal.pone.0102355 (2014). 25. Ru hens eine , B. & Heß, M. Embedding 3D models o biological specimens in PDF publica ions. Mic osc Res Techn 71, 778–786, h ps://doi.o g/10.1002/jem .20618 (2008). 26. Ru hens eine , B., Baeumle , N. & Ba nes, D. In e ac i e 3D olume ende ing in biomedical publica ions. Mic on 41, 886.e1–886. e17, h ps://doi.o g/10.1016/j.mic on.2010.03.010 (2010). 27. Menn, J. & Selige , G. Inc easing Knowledge and Skills o Assembly P ocesses h ough In e ac i e 3D-PDFs. P ocedia CIRP 48, 454–459, h ps://doi.o g/10.1016/j.p oci .2016.02.093 (2016). 28. Semple, T., Peakall, R. & Ta a nic, N. A comp ehensi e and use - iendly amewo k o 3D-da a isualisa ion in in e eb a es and o he o ganisms. J Mo phol 280, 223–231, h ps://doi.o g/10.1002/jmo .20938 (2019). 29. Me sche , B. D. Mic oCT o compa a i e mo phology: simple s aining me hods allow high-con as 3D imaging o di e se non- mine alized animal issues. BMC Physiol 9, 11, h ps://doi.o g/10.1186/1472-6793-9-11 (2009). 30. Me sche , B. D. Mic oCT o de elopmen al biology: A e sa ile ool o high-con as 3D imaging a his ological esolu ions. De Dyn. 238, 632–640, h ps://doi.o g/10.1002/d dy.21857 (2009). 31. Wilkins, S., Gu eye , T., Gao, D., Pogany, A. & S e enson, A. Phase-con as imaging using polych oma ic ha d X- ays. Na u e 384, 335–338, h ps://doi.o g/10.1038/384335a0 (1996). 32. B a an, P. e al. High-Resolu ion X-Ray Phase-Con as 3-D Imaging o B eas Tissue Specimens as a Possible Adjunc o His opa hology. IEEE T ans Med Imaging 37, 2642–2650, h ps://doi.o g/10.1109/TMI.2018.2845905 (2018). 33. Wagne , W. e al. Towa ds synch o on phase-con as lung imaging in pa ien s – a p oo -o -concep s udy on po cine lungs in a human-scale ches phan om. J Synch o on Rad 25, 1827–1832, h ps://doi.o g/10.1107/S1600577518013401 (2018). 34. Momose, A., Takeda, T., I aj, Y. & Hi ano, K. Phase−con as X− ay compu ed omog aphy o obse ing biological so issues. Na . Med. 2 (1996). 35. Saccomano, M. e al. Synch o on inline phase con as µCT enables de ailed i ual his ology o embedded so ‐ issue samples wi h and wi hou s aining. J. Synch o on Radia . 25, h ps://doi.o g/10.1107/S1600577518005489 (2018). 36. La sson, D. H., Vågbe g, W., Ya oshenko, A., Yildi im, A. Ö. & He z, H. M. High- esolu ion sho -exposu e small-animal labo a o y x- ay phase-con as omog aphy. Sci. Rep. 6, h ps://doi.o g/10.1038/s ep39074 (2016). 37. Yang, M. e al. Theo e ical a iance analysis o single- and dual-ene gy compu ed omog aphy me hods o calcula ing p o on s opping powe a ios o biological issues. Phys Med Biol 55, 1343–1362, h ps://doi.o g/10.1088/0031-9155/55/5/006 (2010). 38. Yan, D., Zhang, Z., Luo, Q., Yang, X. & Chen, C. A No el Mouse Segmen a ion Me hod Based on Dynamic Con as Enhanced Mic o-CT Images. PLoS ONE 12, h ps://doi.o g/10.1371/jou nal.pone.0169424 (2017). 39. Lusic, H. & G ins a , M. W. X-Ray Compu ed Tomog aphy Con as Agen s. Chem Re 3, 113, h ps://doi.o g/10.1021/c 200358s (2013). 40. Heude, E., Ri als, I., Couly, G. & Le i, G. Mas ica o y muscle de ec s in hemi acial mic osomia: A new emb yological concep . Am J Med Gene Pa A 155, 1991–1995, h ps://doi.o g/10.1002/ajmg.a.34095 (2011). 41. de Bou non ille, S., Vang unde beeck, S. & Ke ckho s, G. Con as -Enhanced Mic oCT o Vi ual 3D Ana omical Pa hology o Biological Tissues: A Li e a u e Re iew. Con as Media Mol. Imaging 2019, 1–9, h ps://doi.o g/10.1155/2019/8617406 (2019). 42. Zi k mund, T. e al. J. High-con as di e en ia ion esolu ion 3D imaging o oden b ain by X- ay compu ed mic o omog aphy. J Ins um 13, C02039–C02039, h ps://doi.o g/10.1088/1748-0221/13/02/C02039 (2018). 43. Weinha d , V. e al. Quan i a i e mo phome ic analysis o adul eleos ish by X- ay compu ed omog aphy. Sci Rep 8, 16531, h ps://doi.o g/10.1038/s41598-018-34848-z (2018). 44. B o cca di, M. e al. Su ey o P o ocols o he Manual Segmen a ion o he Hippocampus: P epa a o y S eps Towa ds a Join EADC- ADNI Ha monized P o ocol. J Alzheime s Dis 26, 61–75, h ps://doi.o g/10.3233/JAD-2011-0004 (2011). 45. Tes ařo á, M. e al. Use o mic o compu ed- omog aphy and 3D p in ing o e e se enginee ing o mouse emb yo nasal capsule. J Ins um 11, C03006–C03006, h ps://doi.o g/10.1088/1748-0221/11/03/C03006 (2016). 46. Tesařo á, M. e al. A quan i a i e analysis o 3D-cell dis ibu ion in egene a ing muscle-skele al sys em wi h synch o on X- ay compu ed mic o omog aphy. Sci Rep 8, 14145, h ps://doi.o g/10.1038/s41598-018-32459-2 (2018). 47. Cignoni, P. e al. MeshLab: an Open-Sou ce Mesh P ocessing Tool. h p:// cg.is i.cn .i /Publica ions/2008/CCCDGR08/ MeshLabEGIT. inal.pd (2008). 48. Hess, R. The essen ial Blende : guide o 3D c ea ion wi h he open sou ce sui e Blende (ed. Roosendaal, T.) (No S a ch P ess, 2007). 49. 3D PDF Make [so wa e] www.3dpd make .com (2019). 50. Heude, E. e al. Unique mo phogene ic signa u es de ine mammalian neck muscles and associa ed connec i e issues. eLi e 7, e40179, h ps://doi.o g/10.7554/eLi e.40179 (2018). 51. P ochazka, D. e al. Join u iliza ion o double-pulse lase -induced b eakdown spec oscopy and X- ay compu ed omog aphy o olume ic in o ma ion o geological samples. J Anal A om Spec om 33, 1993–1999, h ps://doi.o g/10.1039/c8ja00232k (2018). Acknowledgemen s This esea ch was ca ied ou unde he p ojec CEITEC 2020 (LQ1601) wi h inancial suppo om he Minis y o Educa ion, You h and Spo s o he Czech Republic unde he Na ional Sus ainabili y P og amme II, CEITEC Nano Resea ch In as uc u e (MEYS CR, 2016–2019) and) and Cei ec Nano+ p ojec , CZ.02.01/0.0./.0.0./16 _013/0001728 unde he p og am OP RDE. We acknowledge suppo by g an AFM Tele hon 21853. J.K. and M.T. acknowledge he suppo o he B no Uni e si y o Technology h ough g an FSI-S-17-4506. M.T. was inancially suppo ed by g an CEITEC VUT-J-19-5764 and by he B no Ci y Municipali y as a B no Ph.D. Talen Schola ship Holde . Au ho con ibu ions M.T., E.H. and G.C. wo ked on he da a analysis and me hodology. M.K. p epa ed he samples and s ained hem o CT measu emen . M.T. and T.Z. pe o med CT measu emen s. I.A., S.T. and J.K. designed he expe imen s. M.T., T.Z. and J.K. w o e he manusc ip . All au ho s e iewed he manusc ip . 8 Scien i ic RepoR S | (2019) 9:14896 | h ps://doi.o g/10.1038/s41598-019-51180-2 www.na u e.com/scien i ic epo s www.na u e.com/scien i ic epo s/ compe ing in e es s The au ho s decla e no compe ing in e es s. Addi ional in o ma ion Supplemen a y in o ma ion is a ailable o his pape a h ps://doi.o g/10.1038/s41598-019-51180-2. Co espondence and eques s o ma e ials should be add essed o J.K. Rep in s and pe missions in o ma ion is a ailable a www.na u e.com/ ep in s. Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C e- a i e Commons license, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons license, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons license and you in ended use is no pe - mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his license, isi h p://c ea i ecommons.o g/licenses/by/4.0/. © The Au ho (s) 2019