{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T21:58:40Z","timestamp":1777586320638,"version":"3.51.4"},"reference-count":24,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2025,4,18]],"date-time":"2025-04-18T00:00:00Z","timestamp":1744934400000},"content-version":"vor","delay-in-days":107,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"RAPID-axSpA and C-OPTIMISE studies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,3,14]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>Diagnosis of axial spondyloarthritis (axSpA) is primarily established through the identification of the presence or absence of radiographic sacroiliitis. However, the reliability of conventional radiographs (X-rays) is undermined by significant interreader variability. A machine learning tool could reduce diagnosis time, thereby minimising interreader variability. The present study aimed to evaluate the performance of a deep learning model for detecting radiographic sacroiliitis in axSpA patients from the RAPID-axSpA (NCT01087762) and C-OPTIMISE (NCT02505542) trials.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>Radiographs from the RAPID-axSpA and C-OPTIMISE cohorts were retrospectively used. The deep learning model was previously trained by using a transfer learning approach on non-medical data. The model\u2019s agreement with expert readers was tested on baseline X-rays using central reader data. Sensitivity, specificity, Cohen\u2019s \u03ba, positive and negative predictive values and the area under the receiver operating characteristics curve were calculated.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>The model\u2019s performance was evaluated in the RAPID-axSpA (n\u2009=\u2009277) and C-OPTIMISE (n\u2009=\u2009739) cohorts. In RAPID-axSpA, the model achieved 82% sensitivity, 81% specificity and a Cohen\u2019s \u03ba of 0.61, closely matching central reader performance. In C-OPTIMISE, the model demonstrated 90% sensitivity, 56% specificity and a Cohen\u2019s \u03ba of 0.48. The agreement between the model and central readers was 82% (RAPID-axSpA) and 75% (C-OPTIMISE).<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>The tested deep learning model exhibited accurate radiographic sacroiliitis detection in axSpA patients from diverse clinical trials. The proposed deep learning model could expedite diagnosis, reduce healthcare resource usage and improve patient care pathways.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/rap\/rkae118","type":"journal-article","created":{"date-parts":[[2025,4,18]],"date-time":"2025-04-18T17:01:29Z","timestamp":1744995689000},"source":"Crossref","is-referenced-by-count":3,"title":["Machine learning <i>vs<\/i> human experts: sacroiliitis analysis from the RAPID-axSpA and C-OPTIMISE phase 3 axSpA trials"],"prefix":"10.1093","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4306-033X","authenticated-orcid":false,"given":"Fabian","family":"Proft","sequence":"first","affiliation":[{"name":"Department of Gastroenterology, Infectious Diseases and Rheumatology, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, corporate member of Freie Universit\u00e4t Berlin and Humboldt-Universit\u00e4t zu Berlin , Berlin,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1537-1424","authenticated-orcid":false,"given":"Janis L","family":"Vahldiek","sequence":"additional","affiliation":[{"name":"Department of Radiology, Universit\u00e4tsmedizin Berlin, corporate member of Freie Charit\u00e9 \u2013 Universit\u00e4t Berlin and Humboldt-Universit\u00e4t zu Berlin , Berlin,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2789-8781","authenticated-orcid":false,"given":"Joeri","family":"Nicolaes","sequence":"additional","affiliation":[{"name":"UCB Pharma , Brussels,","place":["Belgium"]},{"name":"Department of Electrical Engineering, Center for Processing Speech and Images, KU Leuven , Leuven,","place":["Belgium"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rachel","family":"Tham","sequence":"additional","affiliation":[{"name":"UCB Pharma , Slough,","place":["UK"]},{"name":"Veramed , London,","place":["UK"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bengt","family":"Hoepken","sequence":"additional","affiliation":[{"name":"UCB Pharma , Monheim am Rhein,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8909-4028","authenticated-orcid":false,"given":"Baran","family":"Ufuktepe","sequence":"additional","affiliation":[{"name":"UCB Pharma , Istanbul,","place":["Turkey"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"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, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, corporate member of Freie Universit\u00e4t Berlin and Humboldt-Universit\u00e4t zu Berlin , Berlin,","place":["Germany"]},{"name":"Department of Epidemiology, German Rheumatism Research Centre , Berlin,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"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, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, corporate member of Freie Universit\u00e4t Berlin and Humboldt-Universit\u00e4t zu Berlin , Berlin,","place":["Germany"]},{"name":"Department of Radiology and Nuclear Medicine, German Heart Center Munich , Munich,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,4,18]]},"reference":[{"key":"2025052701321830200_rkae118-B1","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1002\/art.1780270401","article-title":"Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria","volume":"27","author":"van der Linden","year":"1984","journal-title":"Arthritis Rheum"},{"key":"2025052701321830200_rkae118-B2","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":"Lancet"},{"key":"2025052701321830200_rkae118-B3","doi-asserted-by":"crossref","first-page":"875","DOI":"10.1016\/j.berh.2015.04.031","article-title":"Assessment of spinal pain","volume":"28","author":"Braun","year":"2014","journal-title":"Best Pract Res Clin Rheumatol"},{"key":"2025052701321830200_rkae118-B4","doi-asserted-by":"crossref","first-page":"105468","DOI":"10.1016\/j.jbspin.2022.105468","article-title":"Challenges in non-radiographic axial spondyloarthritis","volume":"90","author":"Poddubnyy","year":"2023","journal-title":"Joint Bone Spine"},{"key":"2025052701321830200_rkae118-B5","doi-asserted-by":"crossref","first-page":"525","DOI":"10.1080\/1744666X.2018.1477591","article-title":"Radiographic progression in non-radiographic axial spondyloarthritis","volume":"14","author":"Protopopov","year":"2018","journal-title":"Expert Rev Clin Immunol"},{"key":"2025052701321830200_rkae118-B6","doi-asserted-by":"crossref","first-page":"1327","DOI":"10.1136\/annrheumdis-2014-206971","article-title":"EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice","volume":"74","author":"Mandl","year":"2015","journal-title":"Ann Rheum Dis"},{"key":"2025052701321830200_rkae118-B7","doi-asserted-by":"crossref","first-page":"791","DOI":"10.1136\/annrheumdis-2015-208852","article-title":"The term \u2018non-radiographic axial spondyloarthritis\u2019 is much more important to classify than to diagnose patients with axial spondyloarthritis","volume":"75","author":"Deodhar","year":"2016","journal-title":"Ann Rheum Dis"},{"key":"2025052701321830200_rkae118-B8","first-page":"125","article-title":"Radiological scoring methods in ankylosing spondylitis. Reliability and change over 1 and 2 years","volume":"31","author":"Spoorenberg","year":"2004","journal-title":"J Rheumatol"},{"key":"2025052701321830200_rkae118-B9","doi-asserted-by":"crossref","first-page":"70","DOI":"10.3899\/jrheum.160079","article-title":"Limited reliability of radiographic assessment of sacroiliac joints in patients with suspected early","volume":"44","author":"Christiansen","year":"2017","journal-title":"J Rheumatol"},{"key":"2025052701321830200_rkae118-B10","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1136\/ard.46.2.139","article-title":"Observer variation in grading sacroiliac radiographs might be a cause of \u2018sacroiliitis\u2019 reported in certain disease states","volume":"46","author":"Yazici","year":"1987","journal-title":"Ann Rheum Dis"},{"key":"2025052701321830200_rkae118-B11","doi-asserted-by":"crossref","first-page":"1369","DOI":"10.1136\/ard.2010.145995","article-title":"Rates and predictors of radiographic sacroiliitis progression over 2 years in patients with axial spondyloarthritis","volume":"70","author":"Poddubnyy","year":"2011","journal-title":"Ann Rheum Dis"},{"key":"2025052701321830200_rkae118-B12","doi-asserted-by":"crossref","first-page":"2403","DOI":"10.1002\/art.38738","article-title":"Agreement between clinical practice and trained central reading in reading of sacroiliac joints on plain pelvic radiographs: results from the Desir cohort","volume":"66","author":"van den Berg","year":"2014","journal-title":"Arthritis Rheumatol"},{"key":"2025052701321830200_rkae118-B13","doi-asserted-by":"crossref","first-page":"106","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":"2025052701321830200_rkae118-B14","doi-asserted-by":"crossref","first-page":"5868","DOI":"10.1093\/rheumatology\/keab636","article-title":"Detection of radiographic sacroiliitis with an artificial neural network in patients with suspicion of axial spondyloarthritis","volume":"60","author":"Poddubnyy","year":"2021","journal-title":"Rheumatology (Oxford)"},{"key":"2025052701321830200_rkae118-B15","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1136\/annrheumdis-2013-204231","article-title":"Efficacy of certolizumab pegol on signs and symptoms of axial spondyloarthritis including ankylosing spondylitis: 24-week results of a double-blind randomised placebo-controlled phase 3 study","volume":"73","author":"Landew\u00e9","year":"2014","journal-title":"Ann Rheum Dis"},{"key":"2025052701321830200_rkae118-B16","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1002\/art.38973","article-title":"Effect of certolizumab pegol over ninety-six weeks in patients with axial spondyloarthritis: results from a phase III randomized trial","volume":"67","author":"Sieper","year":"2015","journal-title":"Arthritis Rheumatol"},{"key":"2025052701321830200_rkae118-B17","doi-asserted-by":"crossref","first-page":"581","DOI":"10.1007\/s40744-020-00214-7","article-title":"Induction of sustained clinical remission in early axial spondyloarthritis following certolizumab pegol treatment: 48-week outcomes from C-OPTIMISE","volume":"7","author":"Landew\u00e9","year":"2020","journal-title":"Rheumatol Ther"},{"key":"2025052701321830200_rkae118-B18","first-page":"8026","article-title":"PyTorch: an imperative style, high performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Adv Neural Inform Process Syst"},{"key":"2025052701321830200_rkae118-B19","doi-asserted-by":"crossref","first-page":"108","DOI":"10.3390\/info11020108","article-title":"Fastai: a layered API for deep learning","volume":"11","author":"Howard","year":"2020","journal-title":"Information"},{"key":"2025052701321830200_rkae118-B20","first-page":"618","author":"Selvaraju","year":"2017"},{"key":"2025052701321830200_rkae118-B21","doi-asserted-by":"crossref","first-page":"1498","DOI":"10.1093\/rheumatology\/kex174","article-title":"Sustained efficacy, safety and patient-reported outcomes of certolizumab pegol in axial spondyloarthritis: 4-year outcomes from RAPID-axSpA","volume":"56","author":"van der Heijde","year":"2017","journal-title":"Rheumatology (Oxford)"},{"key":"2025052701321830200_rkae118-B22","doi-asserted-by":"crossref","first-page":"405","DOI":"10.3348\/kjr.2019.0025","article-title":"Design characteristics of studies reporting the performance of artificial intelligence algorithms for diagnostic analysis of medical images: results from recently published papers","volume":"20","author":"Kim","year":"2019","journal-title":"Korean J Radiol"},{"key":"2025052701321830200_rkae118-B23","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1038\/s41586-019-1799-6","article-title":"International evaluation of an AI system for breast cancer screening","volume":"577","author":"McKinney","year":"2020","journal-title":"Nature"},{"key":"2025052701321830200_rkae118-B24","doi-asserted-by":"crossref","first-page":"398","DOI":"10.1016\/j.oret.2021.12.021","article-title":"Deep learning to detect OCT-derived diabetic macular edema from color retinal photographs: a multicenter validation study","volume":"6","author":"Liu","year":"2022","journal-title":"Ophthalmol Retina"}],"container-title":["Rheumatology Advances in Practice"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/rheumap\/article-pdf\/9\/2\/rkae118\/62945030\/rkae118.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/rheumap\/article-pdf\/9\/2\/rkae118\/62945030\/rkae118.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T05:32:28Z","timestamp":1748323948000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/rheumap\/article\/doi\/10.1093\/rap\/rkae118\/8115332"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":24,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,3,14]]}},"URL":"https:\/\/doi.org\/10.1093\/rap\/rkae118","relation":{},"ISSN":["2514-1775"],"issn-type":[{"value":"2514-1775","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025]]},"published":{"date-parts":[[2025]]},"article-number":"rkae118"}}