{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T15:19:17Z","timestamp":1793459957480,"version":"4.5.0"},"reference-count":33,"publisher":"BMJ","issue":"4","license":[{"start":{"date-parts":[[2024,12,23]],"date-time":"2024-12-23T00:00:00Z","timestamp":1734912000000},"content-version":"unspecified","delay-in-days":22,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Bundesministerium f\u00fcr Bildung und Forschung, BMBF"},{"name":"Abbott"},{"DOI":"10.13039\/501100002347","name":"Federal Ministry of Education and Research","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100002347","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100006483","name":"AbbVie","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006483","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007110","name":"Schering\u2013Plough","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100007110","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100004342","name":"Wyeth","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004342","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100002429","name":"Amgen","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100002429","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100005565","name":"Centocor","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100005565","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["RMD Open"],"accepted":{"date-parts":[[2024,11,5]]},"published-print":{"date-parts":[[2024,12]]},"abstract":"<jats:sec>\n                  <jats:title>Purpose<\/jats:title>\n                  <jats:p>To examine whether incorporating anatomy-centred deep learning can improve generalisability and enable prediction of disease progression.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>This retrospective multicentre study included conventional pelvic radiographs of four different patient cohorts focusing on axial spondyloarthritis collected at university and community hospitals. The first cohort, which consisted of 1483 radiographs, was split into training (n=1261) and validation (n=222) sets. The other cohorts comprising 436, 340 and 163 patients, respectively, were used as independent test datasets. For the second cohort, follow-up data of 311 patients was used to examine progression prediction capabilities. Two neural networks were trained, one on images cropped to the bounding box of the sacroiliac joints (anatomy-centred) and the other one on full radiographs. The performance of the models was compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>On the three test datasets, the standard model achieved AUC scores of 0.853, 0.817, 0.947, with an accuracy of 0.770, 0.724, 0.850. Whereas the anatomy-centred model achieved AUC scores of 0.899, 0.846, 0.957, with an accuracy of 0.821, 0.744, 0.906, respectively. The patients who were identified as high risk by the anatomy-centred model had an OR of 2.16 (95% CI 1.19, 3.86) for having progression of radiographic sacroiliitis within 2 years.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Anatomy-centred deep learning can improve the generalisability of models in detecting radiographic sacroiliitis. The model is published as fully open source alongside this study.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1136\/rmdopen-2024-004628","type":"journal-article","created":{"date-parts":[[2024,12,27]],"date-time":"2024-12-27T10:36:34Z","timestamp":1735295794000},"page":"e004628","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":14,"title":["Anatomy-centred deep learning improves generalisability and progression prediction in radiographic sacroiliitis detection"],"prefix":"10.1136","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-5914-3233","authenticated-orcid":false,"given":"Felix J.","family":"Dorfner","sequence":"first","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1537-1424","authenticated-orcid":false,"given":"Janis L.","family":"Vahldiek","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leonhard","family":"Donle","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrei","family":"Zhukov","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lina","family":"Xu","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hartmut","family":"H\u00e4ntze","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marcus R.","family":"Makowski","sequence":"additional","affiliation":[{"name":"Department of Diagnostic and Interventional Radiology, Technical University of Munich, Munchen, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hugo J. W. L.","family":"Aerts","sequence":"additional","affiliation":[{"name":"Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, Massachusetts, USA"},{"name":"Departments of Radiation Oncology and Radiology, Brigham and Women\u2019s Hospital, Boston, Massachusetts, USA"},{"name":"Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Maastricht, Netherlands"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4306-033X","authenticated-orcid":false,"given":"Fabian","family":"Proft","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5612-043X","authenticated-orcid":false,"given":"Valeria","family":"Rios Rodriguez","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7442-2570","authenticated-orcid":false,"given":"Judith","family":"Rademacher","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4840-5069","authenticated-orcid":false,"given":"Mikhail","family":"Protopopov","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hildrun","family":"Haibel","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kay-Geert","family":"Hermann","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3593-1449","authenticated-orcid":false,"given":"Torsten","family":"Diekhoff","sequence":"additional","affiliation":[{"name":"Department of Radiology, Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lisa C.","family":"Adams","sequence":"additional","affiliation":[{"name":"Department of Diagnostic and Interventional Radiology, Technical University of Munich, Munchen, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4600-9484","authenticated-orcid":false,"given":"Murat","family":"Torgutalp","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4537-6015","authenticated-orcid":false,"given":"Denis","family":"Poddubnyy","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"},{"name":"Division of Rheumatology, University of Toronto, Toronto, Ontario, Canada"},{"name":"Department of Epidemiology, German Rheumatism Research Center, Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9249-8624","authenticated-orcid":false,"given":"Keno K.","family":"Bressem","sequence":"additional","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology (incl. Nutrition Medicine), Charite - Universitatsmedizin Berlin, Berlin, Germany"},{"name":"Department of Diagnostic and Interventional Radiology, Technical University of Munich, Munchen, Germany"},{"name":"Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, Massachusetts, USA"},{"name":"Department of Radiology and Nuclear Medicine, German Heart Center Munich at Technical University Munich, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2024,12,23]]},"reference":[{"key":"2025090312523129000_10.4.e004628.1","doi-asserted-by":"publisher","DOI":"10.1002\/art.1780270401"},{"key":"2025090312523129000_10.4.e004628.2","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1016\/S0140-6736(16)31591-4","article-title":"Axial spondyloarthritis","volume":"390","author":"Sieper","year":"2017","journal-title":"The Lancet"},{"key":"2025090312523129000_10.4.e004628.3","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2014-206971"},{"key":"2025090312523129000_10.4.e004628.4","doi-asserted-by":"publisher","DOI":"10.1136\/ard.46.2.139"},{"key":"2025090312523129000_10.4.e004628.5","doi-asserted-by":"publisher","DOI":"10.1002\/art.38738"},{"key":"2025090312523129000_10.4.e004628.6","doi-asserted-by":"crossref","DOI":"10.1186\/s13075-021-02484-0","article-title":"Deep learning for detection of radiographic sacroiliitis: achieving expert-level performance","volume":"23","author":"Bressem","year":"2021","journal-title":"Arthritis Res Ther"},{"key":"2025090312523129000_10.4.e004628.7","doi-asserted-by":"publisher","DOI":"10.1093\/rheumatology\/keab636"},{"key":"2025090312523129000_10.4.e004628.8","doi-asserted-by":"publisher","DOI":"10.1093\/rheumatology\/keab901"},{"key":"2025090312523129000_10.4.e004628.9","doi-asserted-by":"publisher","DOI":"10.1002\/art.24483"},{"key":"2025090312523129000_10.4.e004628.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.semarthrit.2020.07.018"},{"key":"2025090312523129000_10.4.e004628.11","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2016-210640"},{"key":"2025090312523129000_10.4.e004628.12","doi-asserted-by":"crossref","DOI":"10.1136\/rmdopen-2021-001656","article-title":"CT-like images of the sacroiliac joint generated from MRI using susceptibility-weighted imaging (SWI) in patients with axial spondyloarthritis","volume":"7","author":"Deppe","year":"2021","journal-title":"RMD Open"},{"key":"2025090312523129000_10.4.e004628.13","doi-asserted-by":"publisher","DOI":"10.1136\/annrheumdis-2021-220136"},{"key":"2025090312523129000_10.4.e004628.14","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2012.05.001"},{"key":"2025090312523129000_10.4.e004628.15","doi-asserted-by":"crossref","unstructured":"Diaz-Pinto A , Alle S , Nath V , et al . MONAI label: a framework for ai-assisted interactive labeling of 3d medical images. Med Image Anal [Preprint] 2024. doi:10.1016\/j.media.2024.103207","DOI":"10.1016\/j.media.2024.103207"},{"key":"2025090312523129000_10.4.e004628.16","doi-asserted-by":"crossref","unstructured":"Ronneberger O , Fischer P , Brox T . U-net: convolutional networks for biomedical image segmentation. [Preprint] 2015. Available: https:\/\/doi.org\/10.48550\/arXiv.1505.04597","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2025090312523129000_10.4.e004628.17","unstructured":"Loshchilov I , Hutter F . Decoupled weight decay regularization. [Preprint] 2019. Available: https:\/\/doi.org\/10.48550\/arXiv.1711.05101"},{"key":"2025090312523129000_10.4.e004628.18","doi-asserted-by":"crossref","unstructured":"Smith LN , Topin N . Super-convergence: very fast training of neural networks using large learning rates. In: Pham T , ed. Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications. Baltimore, United States, 2019.","DOI":"10.1117\/12.2520589"},{"key":"2025090312523129000_10.4.e004628.19","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , et al . Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Las Vegas, NV, USA, 2016:770\u20138. doi:10.1109\/CVPR.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"2025090312523129000_10.4.e004628.20","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12172"},{"key":"2025090312523129000_10.4.e004628.21","unstructured":"Zhang H , Cisse M , Dauphin YN , et al . Beyond empirical risk minimization. [Preprint] 2018. Available: https:\/\/doi.org\/10.48550\/arXiv.1710.09412"},{"key":"2025090312523129000_10.4.e004628.22","doi-asserted-by":"crossref","unstructured":"Szegedy C , Vanhoucke V , Ioffe S , et al . Rethinking the inception architecture for computer vision. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Las Vegas, NV, USA. doi:10.1109\/CVPR.2016.308","DOI":"10.1109\/CVPR.2016.308"},{"key":"2025090312523129000_10.4.e004628.23","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-019-01228-7"},{"key":"2025090312523129000_10.4.e004628.24","unstructured":"Falcon W . Pytorch lightning. 2019. Available: https:\/\/doi.org\/10.5281\/zenodo.7469930"},{"key":"2025090312523129000_10.4.e004628.25","unstructured":"Cardoso MJ , Li W , Brown R , et al . Monai: an open-source framework for deep learning in healthcare. Available: https:\/\/doi.org\/10.48550\/arXiv.2211.02701"},{"key":"2025090312523129000_10.4.e004628.26","doi-asserted-by":"publisher","DOI":"10.21105\/joss.04101"},{"key":"2025090312523129000_10.4.e004628.27","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2649-2"},{"key":"2025090312523129000_10.4.e004628.28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2010.04.018"},{"key":"2025090312523129000_10.4.e004628.29","doi-asserted-by":"crossref","unstructured":"McKinney W . Data structures for statistical computing in python. Python in Science Conference; Austin, Texas, 2010:56\u201361. doi:10.25080\/Majora-92bf1922-00a","DOI":"10.25080\/Majora-92bf1922-00a"},{"key":"2025090312523129000_10.4.e004628.30","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"Matplotlib: A 2D Graphics Environment","volume":"9","author":"Hunter","year":"2023","journal-title":"Comput Sci Eng"},{"key":"2025090312523129000_10.4.e004628.31","doi-asserted-by":"publisher","DOI":"10.21105\/joss.03021"},{"key":"2025090312523129000_10.4.e004628.32","doi-asserted-by":"publisher","DOI":"10.1109\/lsp.2014.2337313"},{"key":"2025090312523129000_10.4.e004628.33","doi-asserted-by":"publisher","DOI":"10.1093\/mr\/roab124"}],"container-title":["RMD Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/rmdopen-2024-004628","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T19:52:41Z","timestamp":1756929161000},"score":1,"resource":{"primary":{"URL":"https:\/\/rmdopen.bmj.com\/lookup\/doi\/10.1136\/rmdopen-2024-004628"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12]]},"references-count":33,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,12,23]]},"published-print":{"date-parts":[[2024,12]]}},"alternative-id":["10.1136\/rmdopen-2024-004628"],"URL":"https:\/\/doi.org\/10.1136\/rmdopen-2024-004628","relation":{},"ISSN":["2056-5933"],"issn-type":[{"value":"2056-5933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,12]]}}}