Abstract
Radiographic grading of the sacroiliac joints plays a critical role in the differential diagnosis of ankylosing spondylitis (AS) and in guiding the treatment. The aim of this study was to develop an automated 3D convolutional neural network (3D CNN) for grading and diagnosing sacroiliitis on CT images to assist clinicians. This study included CT images from 2,144 participants, comprising healthy controls and patients with suspected ankylosing spondylitis (AS). A V-Net based segmentation model was applied, followed by training a three-dimensional DenseNet-based convolutional neural network (3D CNN) for both five-class and three-class classification tasks. Grading by three radiologists according to the New York criteria served as the reference standard. The model’s diagnostic performance was evaluated on an external multicenter validation set and compared with radiologist interpretations. For the five-class task, the model’s area under the receiver operating characteristic curve (AUC) for grades 0–IV were 0.966, 0.937, 0.881, 0.962, and 0.994, respectively. In the simplified three-class task, AUCs for classes 0, 1, and 2 were 0.984, 0.967, and 0.994, respectively. On the external validation set, three-class AUCs were 0.957, 0.934, and 0.992. With AI assistance, two radiologists’ diagnostic accuracy improved by 6.9% and 8.4%, respectively. The proposed segmentation–classification framework enables accurate and reproducible CT grading of sacroiliitis.
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on a reasonable request.
References
Agrawal, P., Tote, S. & Sapkale, B. Diagnosis and Treatment of Ankylosing Spondylitis. Cureus 16 (1), e52559. https://doi.org/10.7759/cureus.52559 (2024).
Taurog, J. D., Chhabra, A. & Colbert, R. A. Ankylosing spondylitis and axial spondyloarthritis. N Engl. J. Med. 374, 2563–2574. https://doi.org/10.1056/NEJMra1406182 (2016).
van der Linden, S., Valkenburg, H. A. & Cats, A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum. 27(4), 361-368. https://doi.org/10.1002/art.1780270401 (1984).
van den Berg, R. et al. Agreement between clinical practice and trained central reading in reading of sacroiliac joints on plain pelvic radiographs. Results from the DESIR cohort. Arthritis Rheumatol. 66, 2403–2411 (2014).
Swami, V. G. et al. Diagnostic accuracy of MRI-based sacroiliitis scoring systems: a systematic Review. AJR Am. J. Roentgenol. 12, 1–14 (2019).
Topol, E. J. High-performance medicine: The convergence of human and artificial intelligence. Nat. Med. 25, 44–56 (2019).
Zhang, K. et al. Automatic Image Segmentation and Grading Diagnosis of Sacroiliitis Associated with AS Using a Deep Convolutional Neural Network on CT Images. J. Digit. Imaging. 36 (5), 2025–2034. https://doi.org/10.1007/s10278-023-00858-1 (2023).
Shenkman, Y. et al. Automatic detection and diagnosis of sacroiliitis in CT scans as incidental findings. Med. Image Anal. 57, 165–175. https://doi.org/10.1016/j.media.2019.07.007 (2019).
Liu, L., Zhang, H., Zhang, W., Mei, W. & Huang, R. Sacroiliitis diagnosis based on interpretable features and multi-task learning. Phys. Med. Biol. 69(4). https://doi.org/10.1088/1361-6560/ad2010 (2024).
Lee, J. S., Oh, B. L., Lee, H. Y., Song, Y. W. & Lee, E. Y. Comorbidity, disability, and healthcare expenditure of ankylosing spondylitis in Korea: A population-based study. PLoS One. 13 (2), e0192524. https://doi.org/10.1371/journal.pone.0192524 (2018). Published 2018 Feb 8.
Melchior, J. et al. Radiography, abdominal CT and MRI compared with sacroiliac joint CT in diagnosis of structural sacroiliitis. Eur. J. Radiol. 95, 169–176 (2017).
Poddubnyy, D. et al. Rates and predictors of radiographic sacroiliitis progression over 2 years in patients with axial spondyloarthritis. Ann. Rheum. Dis. 70 (8), 1369–1374. https://doi.org/10.1136/ard.2010.145995 (2011).
van den Berg, R. et al. Agreement between clinical practice and trained central reading in reading of sacroiliac joints on plain pelvic radiographs. Results from the DESIR cohort. Arthritis Rheumatol. 66 (9), 2403–2411. https://doi.org/10.1002/art.38738 (2014).
Inman, R. D. Axial Spondyloarthritis: Current Advances, Future Challenges. J. Rheum. Dis. 28 (2), 55–59. https://doi.org/10.4078/jrd.2021.28.2.55 (2021).
Li, H. et al. Comprehensive AI-assisted tool for ankylosing spondylitis based on multicenter research outperforms human experts. Front. Public. Health. 11, 1063633. https://doi.org/10.3389/fpubh.2023.1063633 (2023).
Maksymowych, W. P. et al. Data-driven definitions for active and structural MRI lesions in the sacroiliac joint in spondyloarthritis and their predictive utility. Rheumatol. (Oxford). 60 (10), 4778–4789. https://doi.org/10.1093/rheumatology/keab099 (2021).
Lambert, R. G. W., Hermann, K. G. A. & Diekhoff, T. Low-dose computed tomography for axial spondyloarthritis: update on use and limitations. Curr. Opin. Rheumatol. 33 (4), 326–332. https://doi.org/10.1097/BOR.0000000000000803 (2021).
Kingsmore, K. M. et al. An introduction to machine learning and analysis of its use in rheumatic diseases. Nat. Rev. Rheumatol. 17 (12), 710–730. https://doi.org/10.1038/s41584-021-00708-w (2021).
Wang, G. et al. A deep-learning pipeline for the diagnosis and discrimination of viral, non-viral and COVID-19 pneumonia from chest X-ray images. Nat. Biomed. Eng. 5 (6), 509–521. https://doi.org/10.1038/s41551-021-00704-1 (2021).
Quanyang, W. et al. Artificial intelligence in lung cancer screening: Detection, classification, prediction, and prognosis. Cancer Med. 13 (7), e7140. https://doi.org/10.1002/cam4.7140 (2024).
Kaya, Y., Akat, E., Yıldırım, S. & Fusion-Brain-Net A Novel Deep Fusion Model for Brain Tumor Classification. Brain Behav. 15 (5), e70520. https://doi.org/10.1002/brb3.70520 (2025).
Li, W. et al. Neural network-based prognostic predictive tool for gastric cardiac cancer: The worldwide retrospective study. BioData Min. 16(1), 21. https://doi.org/10.1186/s13040-023-00335-z (2023).
Chaudhari, M. et al. Graph Neural Network-Based Approaches for Protein Function Prediction. Methods Mol. Biol. 2947, 49–74. https://doi.org/10.1007/978-1-0716-4662-5_3 (2025).
Kanev, G. K. et al. Predicting the target landscape of kinase inhibitors using 3D convolutional neural networks. PLoS Comput. Biol. 19 (9), e1011301. https://doi.org/10.1371/journal.pcbi.1011301 (2023). Published 2023 Sep 5.
Van Den Berghe, T. et al. Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy. Eur. Radiol. 33 (11), 8310–8323. https://doi.org/10.1007/s00330-023-09704-y (2023).
Meng, X. et al. Deep learning for automated grading of radiographic sacroiliitis. Quant. Imaging Med. Surg. 15 (6), 5137–5150. https://doi.org/10.21037/qims-2024-2742 (2025).
Tenório, A. P. M. et al. Radiomic Quantification for MRI Assessment of Sacroiliac Joints of Patients with Spondyloarthritis. J. Digit. Imaging. 35 (1), 29–38. https://doi.org/10.1007/s10278-021-00559-7 (2022).
Canayaz, E., Altikardes, Z. A., Unsal, A., Korkmaz, H. & Gok, M. Development and validation of machine learning algorithms for early detection of ankylosing spondylitis using magnetic resonance images. Technol. Health Care. 33 (3), 1182–1198. https://doi.org/10.1177/09287329241297887 (2025).
Bressem, K. K. et al. Deep Learning Detects Changes Indicative of Axial Spondyloarthritis at MRI of Sacroiliac Joints. Radiology 307 (3), e239007. https://doi.org/10.1148/radiol.239007 (2023).
Bordner, A. et al. A deep learning model for the diagnosis of sacroiliitis according to Assessment of SpondyloArthritis International Society classification criteria with magnetic resonance imaging. Diagn. Interv Imaging. 104 (7–8), 373–383. https://doi.org/10.1016/j.diii.2023.03.008 (2023).
Li, H. et al. A novel semi-supervised learning model based on pelvic radiographs for ankylosing spondylitis diagnosis reduces 90% of annotation cost. Comput. Biol. Med. 184, 109232. https://doi.org/10.1016/j.compbiomed.2024.109232 (2025).
Lin, Y., Chan, S. C. W., Chung, H. Y., Lee, K. H. & Cao, P. A deep neural network for MRI spinal inflammation in axial spondyloarthritis. Eur. Spine J. 33 (11), 4125–4134. https://doi.org/10.1007/s00586-023-08099-0 (2024).
Funding
This work was supported by Shaanxi Provincial Key Research and Development Program (2025SF-YBXM-052), Shaanxi Provincial Traditional Chinese Medicine Scientific Innovation and Capacity Improvement Plan (TZKN-CXPT-04); Research Project of Shaanxi Provincial Health Commission(2021B001); Xi’an Basic Ability Enhancement Program for Scientific and Technological Innovation (24YXYJ0076); Xi’an Municipal Health Commission Research Project (2025yb26).
Author information
Authors and Affiliations
Contributions
Y. D and R.L drafted the paper, C.L, H.G and H.Q contributed to data collection, H.G, D.S and J.Y contributed to analysis, J.C, P.C and Y.L contributed to the entire research implementation.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics statement
Due to the retrospective nature of the study, informed consent was waived. This study adhered to the principles outlined in the Declaration of Helsinki and was registered at the Chinese Clinical Trial Registry under registration number ChiCTR2500096632. Ethical approval was obtained from the Ethics Committee of Xi’an Fifth Hospital with the ethics approval number 2024 (84).
Consent for publication
Any subject who appears in this manuscript has been informed and informed consent to share data has been obtained.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Du, Yk., Liu, R., Guo, H. et al. Automated grading and diagnosis of sacroiliitis on CT images using a 3D convolutional neural network: a multicenter retrospective study. Sci Rep (2026). https://doi.org/10.1038/s41598-026-44911-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-44911-9
