{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T15:48:24Z","timestamp":1789314504894,"version":"build-2803163510"},"reference-count":35,"publisher":"IOP Publishing","issue":"5","license":[{"start":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T00:00:00Z","timestamp":1739750400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/page\/copyright"},{"start":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T00:00:00Z","timestamp":1739750400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["ES 442\/1-1 and RA 2187\/4-1"],"award-info":[{"award-number":["ES 442\/1-1 and RA 2187\/4-1"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2025,3,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>\n                  <jats:italic>Objective.<\/jats:italic> The purpose of this study was to develop a robust deep learning approach trained with a small <jats:italic>in-vivo<\/jats:italic> MRI dataset for multi-label segmentation of all eight carpal bones for therapy planning and wrist dynamic analysis. <jats:italic>Approach.<\/jats:italic> A small dataset of 15 3.0-T MRI scans from five health subjects was employed within this study. The MRI data was variable with respect to the field of view (FOV), wide range of image intensity, and joint pose. A <jats:italic>two-stage<\/jats:italic> segmentation pipeline using modified 3D U-Net was proposed. In the <jats:italic>first stage<\/jats:italic>, a novel architecture, introduced as expansion transfer learning (ETL), cascades the use of a focused region of interest (ROI) cropped around ground truth for pretraining and a subsequent transfer by an expansion to the original FOV for a primary prediction. The bounding box around the ROI generated was utilized in the <jats:italic>second stage<\/jats:italic> for high-accuracy, labeled segmentations of eight carpal bones. Different metrics including dice similarity coefficient (DSC), average surface distance (ASD) and hausdorff distance (HD) were used to evaluate performance between proposed and four state-of-the-art approaches. <jats:italic>Main results.<\/jats:italic> With an average DSC of 87.8\u2009%, an ASD of 0.46\u2009mm, an average HD of 2.42\u2009mm in all datasets (96.1\u2009%, 0.16\u2009mm, 1.38\u2009mm in 12 datasets after exclusion criteria, respectively), the proposed approach showed an overall strongest performance than comparisons. <jats:italic>Significance.<\/jats:italic> To our best knowledge, this is the first CNN-based multi-label segmentation approach for MRI human carpal bones. The ETL introduced in this work improved the ability to localize a small ROI in a large FOV. Overall, the interplay of a <jats:italic>two-stage<\/jats:italic> approach and ETL culminated in convincingly accurate segmentation scores despite a very small amount of image data.<\/jats:p>","DOI":"10.1088\/1361-6560\/adabae","type":"journal-article","created":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T22:54:35Z","timestamp":1737154475000},"page":"055004","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Multi-label segmentation of carpal bones in MRI using expansion transfer learning"],"prefix":"10.1088","volume":"70","author":[{"given":"Stefan","family":"Raith","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matthias","family":"Deitermann","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5840-2115","authenticated-orcid":true,"given":"Tobias","family":"Pankert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8923-4295","authenticated-orcid":true,"given":"Jianzhang","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ali","family":"Modabber","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frank","family":"H\u00f6lzle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frank","family":"Hildebrand","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J\u00f6rg","family":"Eschweiler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2025,2,17]]},"reference":[{"key":"pmbadabaebib1","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1186\/s13018-016-0432-8","article-title":"Sports-related wrist and hand injuries: a review","volume":"11","author":"Avery","year":"2016","journal-title":"J. Orthop. Surg. Res."},{"key":"pmbadabaebib2","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2021.e07211","article-title":"Covid-19 pulmonary consolidations detection in chest x-ray using progressive resizing and transfer learning techniques","volume":"7","author":"Bhatt","year":"2021","journal-title":"Heliyon"},{"key":"pmbadabaebib3","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2021.104776","article-title":"CSR-Net: cross-scale residual network for multi-objective scaphoid fracture segmentation","volume":"137","author":"Chen","year":"2021","journal-title":"Comput. Biol. Med."},{"key":"pmbadabaebib4","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/acb2ab","article-title":"Pairwise attention-enhanced adversarial model for automatic bone segmentation in ct images","volume":"68","author":"Chen","year":"2023","journal-title":"Phys. Med. Biol."},{"key":"pmbadabaebib5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/WiMOB.2019.8923394","article-title":"Aru-net: Research and application for wrist reference bone segmentation","author":"Chen","year":"2019"},{"key":"pmbadabaebib6","first-page":"pp 358","article-title":"S3d-unet: Separable 3d u-net for brain tumor segmentation","author":"Chen","year":"2019"},{"key":"pmbadabaebib7","first-page":"pp 424","article-title":"3D u-net: learning dense volumetric segmentation from sparse annotation","author":"\u00c7i\u00e7ek","year":"2016"},{"key":"pmbadabaebib8","doi-asserted-by":"publisher","first-page":"2096","DOI":"10.1038\/s41467-022-29637-2","article-title":"A fully automatic ai system for tooth and alveolar bone segmentation from cone-beam CT images","volume":"13","author":"Cui","year":"2022","journal-title":"Nat. Commun."},{"key":"pmbadabaebib9","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-34817-6","article-title":"Segmentation of the proximal femur from MR images using deep convolutional neural networks","volume":"8","author":"Deniz","year":"2018","journal-title":"Sci. Rep."},{"key":"pmbadabaebib10","doi-asserted-by":"publisher","first-page":"188","DOI":"10.3390\/life12020188","article-title":"Anatomy, biomechanics and loads of the wrist joint","volume":"12","author":"Eschweiler","year":"2022","journal-title":"Life"},{"key":"pmbadabaebib11","doi-asserted-by":"publisher","first-page":"310","DOI":"10.1177\/0954411916632791","article-title":"Development of a biomechanical model of the wrist joint for patient-specific model guided surgical therapy planning: part 1","volume":"230","author":"Eschweiler","year":"2016a","journal-title":"Proc. Inst. Mech. Eng. H"},{"key":"pmbadabaebib12","doi-asserted-by":"publisher","first-page":"326","DOI":"10.1177\/0954411916635443.","article-title":"A biomechanical model of the wrist joint for patient-specific model guided surgical therapy: part 2","volume":"230","author":"Eschweiler","year":"2016b","journal-title":"Proc. Inst. Mech. Eng. H"},{"key":"pmbadabaebib13","doi-asserted-by":"publisher","first-page":"05","DOI":"10.1038\/s41598-021-89111-9","article-title":"Deep learning methods allow fully automated segmentation of metacarpal bones to quantify volumetric bone mineral density","volume":"11","author":"Folle","year":"2021","journal-title":"Sci. Rep."},{"key":"pmbadabaebib14","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1016\/j.compmedimag.2017.12.003","article-title":"Wrist: a wrist image segmentation toolkit for carpal bone delineation from MRI","volume":"63","author":"Foster","year":"2018","journal-title":"Comput. Med. Imaging Graph."},{"key":"pmbadabaebib15","first-page":"pp 315","article-title":"Deep sparse rectifier neural networks","author":"Glorot","year":"2011"},{"key":"pmbadabaebib16","doi-asserted-by":"publisher","first-page":"2695","DOI":"10.1109\/TBME.2017.2659223","article-title":"Automatic quantification of radiographic wrist joint space width of patients with rheumatoid arthritis","volume":"64","author":"Huo","year":"2017","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"pmbadabaebib17","first-page":"pp 287","article-title":"Brain tumor segmentation and radiomics survival prediction: Contribution to the brats 2017 challenge","author":"Isensee","year":"2018"},{"key":"pmbadabaebib18","doi-asserted-by":"publisher","first-page":"770","DOI":"10.3390\/life12050770","article-title":"Smoothing the undersampled carpal bone model with small volume and large curvature: a feasibility study","volume":"12","author":"Ji","year":"2022","journal-title":"Life"},{"key":"pmbadabaebib19","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1016\/j.media.2016.10.004","article-title":"Efficient multi-scale 3D CNN with fully connected crf for accurate brain lesion segmentation","volume":"36","author":"Kamnitsas","year":"2017","journal-title":"Med. Image Anal."},{"key":"pmbadabaebib20","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1186\/s12880-022-00793-7","article-title":"Transfer learning for medical image classification: a literature review","volume":"22","author":"Kim","year":"2022","journal-title":"BMC med. imaging"},{"key":"pmbadabaebib21","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"pmbadabaebib22","doi-asserted-by":"publisher","first-page":"2007","DOI":"10.1109\/TIP.2011.2146190","article-title":"A level set method for image segmentation in the presence of intensity inhomogeneities with application to MRI","volume":"20","author":"Li","year":"2011","journal-title":"IEEE Trans. Image Process."},{"key":"pmbadabaebib23","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1515\/cdbme-2020-3006","article-title":"A novel combined level set model for automatic mr image segmentation","volume":"6","author":"Li","year":"2020","journal-title":"Curr. Dir. Biomed. Eng."},{"key":"pmbadabaebib24","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2021.106245","article-title":"A novel combined level set model for carpus segmentation from magnetic resonance images with prior knowledge aligned in polar coordinate system","volume":"208","author":"Li","year":"2021","journal-title":"Comput. Methods Programs Biomed."},{"key":"pmbadabaebib25","doi-asserted-by":"publisher","first-page":"1458","DOI":"10.3390\/life12101458","article-title":"Wrist bone motion during flexion-extension and radial-ulnar deviation: an MRI study","volume":"12","author":"Li","year":"2022","journal-title":"Life"},{"key":"pmbadabaebib26","doi-asserted-by":"publisher","first-page":"2663","DOI":"10.1109\/TMI.2018.2845918","article-title":"H-denseunet: hybrid densely connected unet for liver and tumor segmentation from ct volumes","volume":"37","author":"Li","year":"2018","journal-title":"IEEE Trans. Med. Imaging"},{"key":"pmbadabaebib27","doi-asserted-by":"publisher","first-page":"1459","DOI":"10.1016\/j.jhsa.2008.05.035","article-title":"Relationship between the fracture location and the kinematic pattern in scaphoid nonunion","volume":"33","author":"Moritomo","year":"2008","journal-title":"J. Hand Surg."},{"key":"pmbadabaebib28","doi-asserted-by":"publisher","first-page":"1479","DOI":"10.1007\/s11548-022-02830-w","article-title":"Mandible segmentation from CT data for virtual surgical planning using an augmented two-stepped convolutional neural network","volume":"18","author":"Pankert","year":"2023","journal-title":"Int. J. Comput. Assist. Radiol. Surg."},{"key":"pmbadabaebib29","doi-asserted-by":"publisher","first-page":"2089","DOI":"10.1002\/hbm.25348","article-title":"High resolution automated labeling of the hippocampus and amygdala using a 3d convolutional neural network trained on whole brain 700 isotropic 7t mp2rage MRI","volume":"42","author":"Pardoe","year":"2021","journal-title":"Human Brain Mapp."},{"key":"pmbadabaebib30","first-page":"pp 234","article-title":"U-net: Convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"pmbadabaebib31","first-page":"pp 379","article-title":"Tversky loss function for image segmentation using 3D fully convolutional deep networks","volume":"vol 8","author":"Salehi","year":"2017"},{"key":"pmbadabaebib32","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1109\/TMI.2006.891486","article-title":"A review of methods for correction of intensity inhomogeneity in MRI","volume":"26","author":"Vovk","year":"2007","journal-title":"IEEE Trans. Med. Imaging"},{"key":"pmbadabaebib33","doi-asserted-by":"publisher","first-page":"1116","DOI":"10.1016\/j.neuroimage.2006.01.015","article-title":"User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability","volume":"31","author":"Yushkevich","year":"2006","journal-title":"Neuroimage"},{"key":"pmbadabaebib34","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101621","article-title":"Context-guided fully convolutional networks for joint craniomaxillofacial bone segmentation and landmark digitization","volume":"60","author":"Zhang","year":"2020","journal-title":"Med. Image Anal."},{"key":"pmbadabaebib35","doi-asserted-by":"publisher","first-page":"9","DOI":"10.3389\/fendo.2020.00612","article-title":"Automatic vertebral body segmentation based on deep learning of dixon images for bone marrow fat fraction quantification","volume":"11","author":"Zhou","year":"2020","journal-title":"Front. Ind."}],"container-title":["Physics in Medicine &amp; Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T09:31:14Z","timestamp":1739784674000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adabae"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,17]]},"references-count":35,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2025,2,17]]},"published-print":{"date-parts":[[2025,3,2]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/adabae","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,17]]},"assertion":[{"value":"Multi-label segmentation of carpal bones in MRI using expansion transfer learning","name":"article_title","label":"Article Title"},{"value":"Physics in Medicine & Biology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 Institute of Physics and Engineering in Medicine. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2024-06-02","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-01-17","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-02-17","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}