{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T05:02:59Z","timestamp":1789448579548,"version":"build-2803163510"},"reference-count":31,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T00:00:00Z","timestamp":1789430400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>To develop and independently validate a deep learning framework for multiclass differentiation of osteoarthritis (OA), rheumatoid arthritis (RA), gouty arthritis (GA), and normal controls on hand digital radiographs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This retrospective multicenter study included 1,179 hand radiographs from 520 participants at four institutions. A three-center development cohort comprised 1,001 radiographs from 431 participants and was evaluated using patient-level stratified five-fold cross-validation. ResNet50, ResNet101, DenseNet121, and Vision Transformer (ViT) were trained using an identical pipeline. The internally best model, ViT, was retrained on the complete development cohort and evaluated once on an independent fourth-center cohort of 178 radiographs from 89 participants that was held out from training and model selection.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      ViT achieved the numerically highest internal performance, with an accuracy of 0.937\u2009\u00b1\u20090.029 and a macro-\n                      <jats:italic>F<\/jats:italic>\n                      1 score of 0.937\u2009\u00b1\u20090.029. On independent external validation, ViT correctly classified 133 of 178 radiographs, with accuracy of 0.747, macro-precision of 0.756, macro-recall of 0.747, macro-specificity of 0.916, macro-\n                      <jats:italic>F<\/jats:italic>\n                      1 of 0.744, and macro-AUC of 0.917. External class-wise AUCs were 0.948 for RA, 0.873 for OA, 0.888 for GA, and 0.959 for normal controls.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>\n                      The transformer-based model showed strong internal performance and retained useful discriminative ability in an independent external cohort; however, the reduction in external accuracy and macro-\n                      <jats:italic>F<\/jats:italic>\n                      1 indicated limited cross-center transferability. These findings support the feasibility of the proposed approach while emphasizing the need for broader prospective validation and reader-performance studies before clinical deployment.\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1941638","type":"journal-article","created":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T04:27:21Z","timestamp":1789446441000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep learning-based multiclass classification for the diagnosis of hand arthritis on digital radiography: a multicenter study"],"prefix":"10.3389","volume":"13","author":[{"given":"Chao","family":"Deng","sequence":"first","affiliation":[{"name":"Department of Radiology, The First People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dan","family":"Yan","sequence":"additional","affiliation":[{"name":"Department of Radiology, The First People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cong","family":"Tang","sequence":"additional","affiliation":[{"name":"Department of Radiology, Suining Central Hospital","place":["Suining, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuemei","family":"Deng","sequence":"additional","affiliation":[{"name":"Department of Radiology, The Second People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chuanyu","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Radiology, Neijiang Hospital of Traditional Chinese Medicine","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xikui","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Radiology, The First People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lihua","family":"Deng","sequence":"additional","affiliation":[{"name":"Department of Radiology, The First People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhe","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Radiology, The First People\u2019s Hospital of Neijiang","place":["Neijiang, Sichuan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1873","DOI":"10.1016\/s0140-6736(25)01637-x","article-title":"Burden of 375 diseases and injuries, risk-attributable burden of 88 risk factors, and healthy life expectancy in 204 countries and territories, including 660 subnational locations, 1990-2023: a systematic analysis for the global burden of disease study 2023","volume":"406","year":"2025","journal-title":"Lancet"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1300","DOI":"10.1016\/j.joca.2025.08.015","article-title":"Osteoarthritis year in review 2025: epidemiology and therapy","volume":"33","author":"Dell'Isola","year":"2025","journal-title":"Osteoarthritis Cartilage"},{"key":"B3","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1038\/s41572-025-00594-6","article-title":"Osteoarthritis","volume":"11","author":"Tang","year":"2025","journal-title":"Nat Rev Dis Primers"},{"key":"B4","doi-asserted-by":"publisher","first-page":"28","DOI":"10.1016\/j.joca.2025.11.004","article-title":"Osteoarthritis year in review 2025: imaging","volume":"34","author":"Harkey","year":"2026","journal-title":"Osteoarthritis Cartilage"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1659","DOI":"10.1136\/annrheumdis-2013-203355","article-title":"Incidence and risk factors for clinically diagnosed knee, hip and hand osteoarthritis: influences of age, gender and osteoarthritis affecting other joints","volume":"73","author":"Prieto-Alhambra","year":"2014","journal-title":"Ann Rheum Dis"},{"key":"B6","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s41413-025-00423-2","article-title":"Advancing osteoarthritis research: the role of AI in clinical, imaging and omics fields","volume":"13","author":"Ou","year":"2025","journal-title":"Bone Res"},{"key":"B7","doi-asserted-by":"publisher","first-page":"100161","DOI":"10.1016\/j.ostima.2023.100161","article-title":"Artificial intelligence in knee osteoarthritis: a comprehensive review for 2022","volume":"3","author":"Cigdem","year":"2023","journal-title":"Osteoarthr Imaging"},{"key":"B8","doi-asserted-by":"publisher","first-page":"3032","DOI":"10.1002\/ksa.12702","article-title":"Artificial intelligence-assisted analysis of musculoskeletal imaging-a narrative review of the current state of machine learning models","volume":"33","author":"Oettl","year":"2025","journal-title":"Knee Surg Sports Traumatol Arthrosc"},{"key":"B9","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1016\/j.joca.2023.10.005","article-title":"Osteoarthritis year in review 2023: imaging","volume":"32","author":"Jarraya","year":"2024","journal-title":"Osteoarthritis Cartilage"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1289","DOI":"10.1007\/s10278-025-01579-3","article-title":"Radiographic bone texture analysis using deep learning models for early rheumatoid arthritis diagnosis","volume":"39","author":"Huang","year":"2025","journal-title":"J Imaging Inform Med"},{"key":"B11","doi-asserted-by":"publisher","first-page":"3391","DOI":"10.1038\/s41598-025-86073-0","article-title":"Multistage deep learning methods for automating radiographic sharp score prediction in rheumatoid arthritis","volume":"15","author":"Moradmand","year":"2025","journal-title":"Sci Rep"},{"key":"B12","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1186\/s12880-024-01362-w","article-title":"Radiograph-based rheumatoid arthritis diagnosis via convolutional neural network","volume":"24","author":"Peng","year":"2024","journal-title":"BMC Med Imaging"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1333","DOI":"10.2147\/itt.S547159","article-title":"Deep learning classification of rheumatoid arthritis in hand radiographs interpretability insights and web application","volume":"14","author":"Cai","year":"2025","journal-title":"Immunotargets Ther"},{"key":"B14","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1007\/s10278-021-00564-w","article-title":"Automated classification of rheumatoid arthritis, osteoarthritis, and normal hand radiographs with deep learning methods","volume":"35","author":"\u00dcreten","year":"2022","journal-title":"J Digit Imaging"},{"key":"B15","doi-asserted-by":"publisher","first-page":"377","DOI":"10.1007\/s00256-023-04408-2","article-title":"Deep learning discrimination of rheumatoid arthritis from osteoarthritis on hand radiography","volume":"53","author":"Ma","year":"2024","journal-title":"Skeletal Radiol"},{"key":"B16","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1002\/acr2.11665","article-title":"Detection and grading of radiographic hand osteoarthritis using an automated machine learning platform","volume":"6","author":"Caratsch","year":"2024","journal-title":"ACR Open Rheumatol"},{"key":"B17","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"B18","first-page":"2261","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"B19","article-title":"An image is worth 16\u00d716 words: transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"B20","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1186\/s41747-024-00422-8","article-title":"AI applications in musculoskeletal imaging: a narrative review","volume":"8","author":"Gitto","year":"2024","journal-title":"Eur Radiol Exp"},{"key":"B21","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1007\/s00256-021-03909-2","article-title":"AI MSK clinical applications: cartilage and osteoarthritis","volume":"51","author":"Joseph","year":"2022","journal-title":"Skeletal Radiol"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1003","DOI":"10.1016\/j.joca.2023.03.005","article-title":"Osteoarthritis year in review 2022: imaging","volume":"31","author":"Demehri","year":"2023","journal-title":"Osteoarthritis Cartilage"},{"key":"B23","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1016\/j.joca.2024.10.009","article-title":"Osteoarthritis year in review 2024: imaging","volume":"33","author":"Hayashi","year":"2025","journal-title":"Osteoarthritis Cartilage"},{"key":"B24","doi-asserted-by":"publisher","first-page":"e001998","DOI":"10.1136\/rmdopen-2021-001998","article-title":"Use of machine learning in osteoarthritis research: a systematic literature review","volume":"8","author":"Binvignat","year":"2022","journal-title":"RMD Open"},{"key":"B25","doi-asserted-by":"publisher","first-page":"152750","DOI":"10.1016\/j.semarthrit.2025.152750","article-title":"Algorithmic approaches in hand imaging for rheumatic musculoskeletal diseases: a systematic literature review","volume":"73","author":"Perronne","year":"2025","journal-title":"Semin Arthritis Rheum"},{"key":"B26","doi-asserted-by":"publisher","first-page":"102721","DOI":"10.1016\/j.media.2022.102721","article-title":"Explaining the black-box smoothly-a counterfactual approach","volume":"84","author":"Singla","year":"2023","journal-title":"Med Image Anal"},{"key":"B27","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1186\/s12880-024-01339-9","article-title":"A deep learning-powered diagnostic model for acute pancreatitis","volume":"24","author":"Zhang","year":"2024","journal-title":"BMC Med Imaging"},{"key":"B28","doi-asserted-by":"publisher","first-page":"110787","DOI":"10.1016\/j.ejrad.2023.110787","article-title":"Explainable AI in medical imaging: an overview for clinical practitioners - saliency-based XAI approaches","volume":"162","author":"Borys","year":"2023","journal-title":"Eur J Radiol"},{"key":"B29","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1186\/s12880-025-02118-w","article-title":"Explainable artificial intelligence (XAI) in medical imaging: a systematic review of techniques, applications, and challenges","volume":"26","author":"Ahmed","year":"2026","journal-title":"BMC Med Imaging"},{"key":"B30","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1016\/j.joca.2023.09.011","article-title":"Artificial intelligence in osteoarthritis detection: a systematic review and meta-analysis","volume":"32","author":"Mohammadi","year":"2024","journal-title":"Osteoarthritis Cartilage"},{"key":"B31","doi-asserted-by":"publisher","first-page":"542","DOI":"10.1016\/j.csbj.2024.08.005","article-title":"Unveiling the black box: a systematic review of explainable artificial intelligence in medical image analysis","volume":"24","author":"Muhammad","year":"2024","journal-title":"Comput Struct Biotechnol J"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1941638\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T04:27:22Z","timestamp":1789446442000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1941638\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,15]]},"references-count":31,"alternative-id":["10.3389\/fmed.2026.1941638"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1941638","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,15]]},"article-number":"1941638"}}