MRI-Based Framework for Ankylosing Spondylitis Diagnosis
DOI:
https://doi.org/10.68337/cpsm.v1.i1.2026-007Keywords:
Ankylosing spondylitis, MRI, deep learning, Attention U-Net, hybrid CNN-Transformer, medical image segmentation, stage-wise classification, explainable AI, Grad-CAMAbstract
Ankylosing spondylitis (AS) is a progressive inflammatory disease of the sacroiliac joints and spine, and its early inflammation can be detected only with magnetic resonance imaging (MRI). Manual detection of early inflammation on MRI is time-consuming, inconsistent, and subjective, which delays diagnosis and treatment. This paper presents an automated approach to detecting AS from MRI scans. The approach segments and classifies MRI images automatically and makes the classification more interpretable. The framework combines several deep learning models. An Attention U-Net segments the sacroiliac joints in the MRI scans. A hybrid convolutional neural network (CNN) and Transformer architecture then classifies each image as AS-positive or AS-negative and attempts to assign the disease stage (normal, early, moderate, or advanced). This architecture uses EfficientNet to extract detailed local features and Vision Transformer blocks to capture global context. Finally, gradient-weighted class activation mapping (Grad-CAM) gives a visual explanation of each decision. On 962 test images, the framework detected AS with 88.67% accuracy; stage grading remained limited, with 60.40% stage accuracy, and most early and moderate cases were predicted as advanced.
References
[1] Mohit Kumar, Piyush Patil, Yash Singh, and Tanmayee Patki, "Early diagnosis of ankylosing spondylitis using MRI and transfer learning," Grenze Int. J. Eng. Technol., vol. 11, no. 1, 2025. [Online]. Available: https://thegrenze.com (GIJET, vol. 11, no. 1, paper 506).
[2] Sıtkı Kocaoğlu, "FPGA implementation of deep learning architecture for ankylosing spondylitis detection from MRI," Sci. Rep., vol. 15, Art. no. 21854, 2025, doi: 10.1038/s41598-025-08593-z.
[3] Keno K. Bressem, Lisa C. Adams, Fabian Proft, Kay Geert A. Hermann, Torsten Diekhoff, Laura Spiller, Stefan M. Niehues, Marcus R. Makowski, Bernd Hamm, Mikhail Protopopov, Valeria Rios Rodriguez, Hildrun Haibel, Judith Rademacher, Murat Torgutalp, Robert G. Lambert, Xenofon Baraliakos, Walter P. Maksymowych, Janis L. Vahldiek, and Denis Poddubnyy, "Deep learning detects changes indicative of axial spondyloarthritis at MRI of sacroiliac joints," Radiology, vol. 305, no. 3, pp. 655-665, 2022, doi: 10.1148/radiol.212526.
[4] Shaik Hasane Ahammad, R. Jayaraj, S. Shibu, V. Sujatha, Ch. Prathima, L. Megalan Leo, R. Thandaiah Prabu, Md. Amzad Hossain, and Ahmed Nabih Zaki Rashed, "Advanced model based machine learning technique for early stage prediction of ankylosing spondylitis under timely analysis with featured textures," Multimed. Tools Appl., vol. 83, no. 26, pp. 68393-68413, 2024, doi: 10.1007/s11042-024-18236-6.
[5] Emre Canayaz, Zehra Aysun Altikardes, Alparslan Unsal, Hayriye Korkmaz, and Mustafa Gok, "Development and validation of machine learning algorithms for early detection of ankylosing spondylitis using magnetic resonance images," Technol. Health Care, vol. 33, no. 3, pp. 1182-1198, 2025, doi: 10.1177/09287329241297887.
[6] Yingying Lin, Shirley Chiu Wai Chan, Ho Yin Chung, Kam Ho Lee, and Peng Cao, "A deep neural network for MRI spinal inflammation in axial spondyloarthritis," Eur. Spine J., vol. 33, no. 11, pp. 4125-4134, 2024, doi: 10.1007/s00586-023-08099-0.
[7] Jonathan Kennedy, Natasha Kennedy, Roxanne Cooksey, Ernest Choy, Stefan Siebert, Muhammad Rahman, and Sinead Brophy, "Predicting a diagnosis of ankylosing spondylitis using primary care health records: A machine learning approach," PLoS One, vol. 18, no. 3, Art. no. e0279076, 2023, doi: 10.1371/journal.pone.0279076.
[8] Pauley Chea and Jacob C. Mandell, "Current applications and future directions of deep learning in musculoskeletal radiology," Skeletal Radiol., vol. 49, no. 2, pp. 183-197, 2020, doi: 10.1007/s00256-019-03284-z.
[9] Xin Li, Yi Lin, Zhuoyao Xie, Zixiao Lu, Liwen Song, Qiang Ye, Menghong Wang, Xiao Fang, Yi He, Hao Chen, and Yinghua Zhao, "Automatic segmentation of fat metaplasia on sacroiliac joint MRI using deep learning," Insights Imaging, vol. 15, Art. no. 93, 2024, doi: 10.1186/s13244-024-01659-y.
[10] Mahmud Omar, Abdulla Watad, Dennis McGonagle, Shelly Soffer, Benjamin S. Glicksberg, Girish N. Nadkarni, and Eyal Klang, "The role of deep learning in diagnostic imaging of spondyloarthropathies: A systematic review," Eur. Radiol., vol. 35, no. 6, pp. 3661-3672, 2025, doi: 10.1007/s00330-024-11261-x.
[11] Fahad Shamshad, Salman Khan, Syed Waqas Zamir, Muhammad Haris Khan, Munawar Hayat, Fahad Shahbaz Khan, and Huazhu Fu, "Transformers in medical imaging: A survey," Med. Image Anal., vol. 88, Art. no. 102802, 2023, doi: 10.1016/j.media.2023.102802.
[12] Qin Wei, Wei Huang, Yanling Chen, Yi Zhang, Kebing Yu, and Shaoping Wu, "Retrospective comparative study of lumbar spine MRI texture analysis in diagnosing bone marrow edema lesions in ankylosing spondylitis and non-ankylosing spondylitis," BMC Musculoskelet. Disord., vol. 26, Art. no. 163, 2025, doi: 10.1186/s12891-025-08413-5.
[13] Zhaojuan Zhang, Yiling Pan, Yanjie Lu, Lusi Ye, Mo Zheng, Guodao Zhang, and Dan Chen, "The TabNet model for diagnosing axial spondyloarthritis using MRI imaging findings and clinical risk factors," Int. J. Rheum. Dis., vol. 27, no. 12, Art. no. e70004, 2024, doi: 10.1111/1756-185X.70004.
[14] Omer Dor, Oz Haim, Lee Azolai, Ariel Agur, May Cohen, Noam Shomron, and Moran Artzi, "Automatic detection of spinal pathologies based on MRI scans," Eur. Spine J., vol. 35, no. 3, pp. 1345-1353, 2026, doi: 10.1007/s00586-025-09547-9.
[15] Qian He, Jing Luo, Jiaqi Chen, Jianying Yang, Chuanhui Yao, Caiqin Xu, and Qingwen Tao, "The validity and reliability of quality of life questionnaires in patients with ankylosing spondylitis and non-radiographic axial spondyloarthritis: A systematic review and meta-analysis," Health Qual. Life Outcomes, vol. 20, Art. no. 116, 2022, doi: 10.1186/s12955-022-02026-5.
[16] Parv Agrawal, Sachin Tote, and Bhagyesh Sapkale, "Diagnosis and treatment of ankylosing spondylitis," Cureus, vol. 16, no. 1, Art. no. e52559, 2024, doi: 10.7759/cureus.52559.
[17] Seulkee Lee, Uju Jeon, Ji Hyun Lee, Seonyoung Kang, Hyungjin Kim, Jaejoon Lee, Myung Jin Chung, and Hoon-Suk Cha, "Artificial intelligence for the detection of sacroiliitis on magnetic resonance imaging in patients with axial spondyloarthritis," Front. Immunol., vol. 14, Art. no. 1278247, 2023, doi: 10.3389/fimmu.2023.1278247.
[18] Andreea-Iulia Nicoara, Lorena-Mihaela Sas, Cristina Elena Bita, Stefan Cristian Dinescu, and Florentin Ananu Vreju, "Implementation of artificial intelligence models in magnetic resonance imaging with focus on diagnosis of rheumatoid arthritis and axial spondyloarthritis: Narrative review," Front. Med., vol. 10, Art. no. 1280266, 2023, doi: 10.3389/fmed.2023.1280266.
[19] Yan Zheng, Chao Bai, Kui Zhang, Qing Han, Qingbiao Guan, Ying Liu, Zhaohui Zheng, Yong Xia, and Ping Zhu, "Deep-learning based quantification model for hip bone marrow edema and synovitis in patients with spondyloarthritis based on magnetic resonance images," Front. Physiol., vol. 14, Art. no. 1132214, 2023, doi: 10.3389/fphys.2023.1132214.
[20] Wenxin Gao, Ruirui Hou, Yungang Chen, Xiaoying Wang, Guoyan Liu, Wanli Hu, Kang Yao, and Yanke Hao, "A predictive disease risk model for ankylosing spondylitis: Based on integrated bioinformatic analysis and identification of potential biomarkers most related to immunity," Mediators Inflamm., vol. 2023, Art. no. 3220235, 2023, doi: 10.1155/2023/3220235.
[21] Nevsun Pihtili Tas, Oguz Kaya, Gulay Macin, Burak Tasci, Sengul Dogan, and Turker Tuncer, "ASNET: A novel AI framework for accurate ankylosing spondylitis diagnosis from MRI," Biomedicines, vol. 11, no. 9, Art. no. 2441, 2023, doi: 10.3390/biomedicines11092441.
[22] Zhuoyao Xie, Zefeiyun Chen, Qinmei Yang, Qiang Ye, Xin Li, Qiuxia Xie, Caolin Liu, Bomiao Lin, Xinai Han, Yi He, Xiaohong Wang, Wei Yang, and Yinghua Zhao, "Enhanced diagnosis of axial spondyloarthritis using machine learning with sacroiliac joint MRI: A multicenter study," Insights Imaging, vol. 16, Art. no. 91, 2025, doi: 10.1186/s13244-025-01967-x.
[23] Riel Castro-Zunti, Eun Hae Park, Hae Ni Park, Younhee Choi, Gong Yong Jin, Hee Suk Chae, and Seok-bum Ko, "Diagnosing ankylosing spondylitis via architecture-modified ResNet and combined conventional magnetic resonance imagery," J. Imaging Inform. Med., vol. 38, no. 6, pp. 3665-3683, 2025, doi: 10.1007/s10278-025-01427-4.
Downloads
Published
Versions
- 2026-10-01 (2)
- 2026-09-30 (1)
Data Availability Statement
This study used publicly available sacroiliac joint MRI images from public medical imaging repositories, including slices from the LSD, SPIDER, OSF, and T-SEG datasets.Conference Proceedings Volume
Section
Categories
License
Copyright (c) 2026 The Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in Conference Proceedings in Science and Management are made available under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Authors retain copyright in their work and grant the journal first publication rights. Readers are free to share and adapt the published material for any purpose, including commercial use, provided appropriate credit is given to the author, a link to the license is provided, and any changes are indicated.
