{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T08:03:39Z","timestamp":1773216219513,"version":"3.50.1"},"reference-count":25,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T00:00:00Z","timestamp":1773100800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JCM"],"abstract":"<jats:p>Background and Objective: Axial spondyloarthritis (axSpA) is a group of chronic inflammatory diseases that primarily affect the sacroiliac joints. Early diagnosis is crucial for preventing irreversible structural damage. Magnetic Resonance Imaging (MRI) is the gold standard for detecting early inflammatory changes such as sacroiliitis. However, conventional MRI interpretation is inherently subjective and susceptible to both intra- and inter-observer variability. Therefore, artificial intelligence (AI)-driven diagnostic solutions are increasingly being explored. Among them, the Gated Attention Multiple Instance Learning (MIL) framework holds strong potential in modeling heterogeneous inflammatory distributions, thanks to its slice-level attention mechanism. This study aims to evaluate the diagnostic performance of a deep learning model based on Gated Attention MIL for automated sacroiliitis detection. Furthermore, its results are compared with a baseline deep learning architecture (standard ResNet-18), and its consistency with radiologist annotations is analyzed. Materials and Methods: The dataset included 554 subjects, comprising 276 patients diagnosed with axSpA and 278 healthy controls. All MRI data were derived from axial T2-weighted fat-suppressed (T2_TSE_TRA_FS) sequences. Patient-wise data splitting was employed to construct training, validation, and independent test sets. The proposed model architecture integrates ResNet-18-based feature extraction, a gated attention mechanism for instance-level weighting, and bag-level classification. Additionally, Test-Time Augmentation (TTA) was implemented to enhance robustness during inference. Results: On the independent test set, the model achieved an accuracy of 85.88%, sensitivity of 92.86%, specificity of 79.07%, and an F1-score of 86.67%. Attention heatmaps generated by the MIL module showed strong spatial overlap with bone marrow edema regions annotated by expert radiologists. Implementation of TTA led to an approximate 10% improvement in overall classification accuracy. Conclusions: The Gated Attention MIL framework demonstrated high diagnostic performance for sacroiliitis detection, indicating its value as a reliable decision support tool for early axSpA diagnosis. Validation on larger, multi-center datasets is warranted to ensure generalizability and to support clinical integration in routine radiology workflows.<\/jats:p>","DOI":"10.3390\/jcm15062101","type":"journal-article","created":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T09:55:38Z","timestamp":1773136538000},"page":"2101","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Gated Attention-Based Multiple Instance Learning and Test-Time Augmentation Approach for Diagnosing Active Sacroiliitis in Sacroiliac Joint MRI Scans"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6496-8748","authenticated-orcid":false,"given":"Zeynep","family":"Keskin","sequence":"first","affiliation":[{"name":"Department of Radiology, Konya City Hospital, Konya 42020, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4573-7025","authenticated-orcid":false,"given":"Onur","family":"\u0130nan","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Technology, Sel\u00e7uk University, Konya 42250, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4004-4991","authenticated-orcid":false,"given":"\u00d6mer","family":"\u00d6zberk","sequence":"additional","affiliation":[{"name":"Department of Radiology, Konya City Hospital, Konya 42020, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2523-1695","authenticated-orcid":false,"given":"Reyhan","family":"Bilici","sequence":"additional","affiliation":[{"name":"Department of Rheumatology, Faculty of Medicine, Necmettin Erbakan University, Konya 42090, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2069-9085","authenticated-orcid":false,"given":"Sema","family":"Servi","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Technology, Sel\u00e7uk University, Konya 42250, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-8546-0070","authenticated-orcid":false,"given":"Selma \u00d6zlem","family":"\u00c7elikdelen","sequence":"additional","affiliation":[{"name":"Department of Internal Medicine, Konya City Hospital, Konya 42020, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mehmet","family":"Y\u0131ld\u0131r\u0131m","sequence":"additional","affiliation":[{"name":"Department of Internal Medicine, Konya City Hospital, Konya 42020, Turkey"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,10]]},"reference":[{"key":"ref_1","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":"ref_2","doi-asserted-by":"crossref","first-page":"e547","DOI":"10.1097\/RHU.0000000000001575","article-title":"Clinical manifestations and diagnosis of axial spondyloarthritis","volume":"27","author":"Walsh","year":"2021","journal-title":"JCR J. Clin. Rheumatol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1136\/ard.2009.108233","article-title":"The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): Validation and final selection","volume":"68","author":"Rudwaleit","year":"2009","journal-title":"Ann. Rheum. Dis."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"2016","DOI":"10.1136\/annrheumdis-2014-205432","article-title":"Classification of axial SpA based on positive imaging (radiographs and\/or MRI of the sacroiliac joints) by local rheumatologists or radiologists versus central trained readers in the DESIR cohort","volume":"74","author":"Lenczner","year":"2015","journal-title":"Ann. Rheum. Dis."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"109755","DOI":"10.1016\/j.ejrad.2021.109755","article-title":"Differentiation of inflammatory from degenerative changes in the sacroiliac joints by machine learning supported texture analysis","volume":"140","author":"Kepp","year":"2021","journal-title":"Eur. J. Radiol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1186\/s42358-020-00126-8","article-title":"Machine learning techniques for computer-aided classification of active inflammatory sacroiliitis in magnetic resonance imaging","volume":"60","author":"Faleiros","year":"2020","journal-title":"Adv. Rheumatol."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"373","DOI":"10.1016\/j.diii.2023.03.008","article-title":"A deep learning model for the diagnosis of sacroiliitis according to Assessment of SpondyloArthritis International Society classification criteria with magnetic resonance imaging","volume":"104","author":"Bordner","year":"2023","journal-title":"Diagn. Interv. Imaging"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"8712","DOI":"10.1109\/JIOT.2025.3641888","article-title":"PHH-FL: Perceptual Hashing Hypernetwork Personalized Federated Learning for Heterogeneous Medical Image Analysis Tasks","volume":"13","author":"Sun","year":"2025","journal-title":"IEEE Internet Things J."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"103819","DOI":"10.1016\/j.media.2025.103819","article-title":"Knowledge distillation and teacher-student learning in medical imaging: Comprehensive overview, pivotal role, and future directions","volume":"107","author":"Li","year":"2025","journal-title":"Med. Image Anal."},{"key":"ref_10","unstructured":"Ilse, M., Tomczak, J., and Welling, M. (2018). Attention-based deep multiple instance learning. International Conference on Machine Learning, PMLR."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Kim, H.E., Cosa-Linan, A., Santhanam, N., Jannesari, M., Maros, M.E., and Ganslandt, T. (2022). Transfer learning for medical image classification: A literature review. BMC Med. Imaging, 22.","DOI":"10.1186\/s12880-022-00793-7"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"101673","DOI":"10.1016\/j.compmedimag.2019.101673","article-title":"Convolutional neural networks for multi-class brain disease detection using MRI images","volume":"78","author":"Talo","year":"2019","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"20311","DOI":"10.1007\/s10489-023-04458-y","article-title":"Triple-kernel gated attention-based multiple instance learning with contrastive learning for medical image analysis","volume":"53","author":"Hu","year":"2023","journal-title":"Appl. Intell."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Ilse, M., Tomczak, J.M., and Welling, M. (2020). Deep multiple instance learning for digital histopathology. Handbook of Medical Image Computing and Computer Assisted Intervention, Academic Press.","DOI":"10.1016\/B978-0-12-816176-0.00027-2"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1016\/S0004-3702(96)00034-3","article-title":"Solving the multiple instance problem with axis-parallel rectangles","volume":"89","author":"Dietterich","year":"1997","journal-title":"Artif. Intell."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"34","DOI":"10.1016\/j.neucom.2019.01.103","article-title":"Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks","volume":"338","author":"Wang","year":"2019","journal-title":"Neurocomputing"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"33","DOI":"10.1007\/s13755-021-00163-7","article-title":"Improving convolutional neural networks performance for image classification using test time augmentation: A case study using MURA dataset","volume":"9","author":"Kandel","year":"2021","journal-title":"Health Inf. Sci. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1136\/annrheumdis-2022-eular.1416","article-title":"OP0152 A deep learn\u0131ng framework for mr\u0131 detect\u0131on of act\u0131ve \u0131nflammatory and structural changes \u0131n the sacro\u0131l\u0131ac jo\u0131nt cons\u0131stent w\u0131th ax\u0131al spondyloarthr\u0131t\u0131s: An \u0131nternat\u0131onal collaborat\u0131ve study","volume":"81","author":"Bressem","year":"2022","journal-title":"Ann. Rheum. Dis."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"111347","DOI":"10.1016\/j.ejrad.2024.111347","article-title":"Use of MRI-based deep learning radiomics to diagnose sacroiliitis related to axial spondyloarthritis","volume":"172","author":"Zhang","year":"2024","journal-title":"Eur. J. Radiol."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"655","DOI":"10.1148\/radiol.212526","article-title":"Deep learning detects changes indicative of axial spondyloarthritis at MRI of sacroiliac joints","volume":"305","author":"Bressem","year":"2022","journal-title":"Radiology"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1002\/jmri.29731","article-title":"Diagnosis of sacroiliitis through semi-supervised segmentation and radiomics feature analysis of MRI images","volume":"62","author":"Liu","year":"2025","journal-title":"J. Magn. Reson. Imaging"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1136\/ard-2024-225862","article-title":"Performance analysis of a deep-learning algorithm to detect the presence of inflammation in MRI of sacroiliac joints in patients with axial spondyloarthritis","volume":"84","author":"Nicolaes","year":"2025","journal-title":"Ann. Rheum. Dis."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2169","DOI":"10.1002\/art.42650","article-title":"Machine Learning Pipeline for Predicting Bone Marrow Edema Along the Sacroiliac Joints on Magnetic Resonance Imaging","volume":"75","author":"Roels","year":"2023","journal-title":"Arthritis Rheumatol."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Nazzal, W., Thurnhofer-Hemsi, K., and L\u00f3pez-Rubio, E. (2024). Improving medical image segmentation using test-time augmentation with medsam. Mathematics, 12.","DOI":"10.3390\/math12244003"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"5068","DOI":"10.1038\/s41598-020-61808-3","article-title":"Test-time augmentation for deep learning-based cell segmentation on microscopy images","volume":"10","author":"Moshkov","year":"2020","journal-title":"Sci. Rep."}],"container-title":["Journal of Clinical Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/6\/2101\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T10:12:25Z","timestamp":1773137545000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/6\/2101"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,10]]},"references-count":25,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["jcm15062101"],"URL":"https:\/\/doi.org\/10.3390\/jcm15062101","relation":{},"ISSN":["2077-0383"],"issn-type":[{"value":"2077-0383","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,10]]}}}