ORIGINAL RESEARCH article
Front. Med.
Sec. Rheumatology
Volume 12 - 2025 | doi: 10.3389/fmed.2025.1620716
Construction of a grading model for sacroiliac joint lesions in ankylosing spondylitis based on CT images
Provisionally accepted- 1Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China
- 2Beijing University of Technology Faculty of Information Science and Technology, Beijing, China., Beijing, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Objective: Constructing an AS sacroiliac joint lesions grading model based on ResNet series network using sacroiliac joint CT images to assist in the early diagnosis and treatment of AS.Methods: The sacroiliac joint CT image data of AS patients who met the inclusion criteria was collected from September 2016 to August 2022 in Guang 'anmen Hospital, China Academy of Chinese Medical Sciences. The data was divided into the training set and the test set according to the ratio of 8:2. The ResNet series network was used to construct the AS sacroiliac joint disease classification model, which was divided into three steps: image preprocessing, task evaluation and hierarchical structure design. The grading results were obtained through the above steps, which were finally divided into 5 levels: 0 level, 1 level, 2 level, 3 level and 4 level.Results: A total of 310 AS patients with complete data were included, which were divided into the training set (247) and the test set (63). The final grading results were consistent with the current clinical grading standards, and the sacroiliac joint lesions of AS could be divided into five grades. The classification performance reached a high level, in which the accuracy was 0.863
Keywords: ankylosing spondylitis, machine learning, CT image, Sacroiliac Joint, Grading model
Received: 30 Apr 2025; Accepted: 04 Aug 2025.
Copyright: © 2025 Sun, Chen, Yang, Zhuo, Xu, Ge, Gong, Yang, Qu, Li and Liu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Xiaoguang Li, Beijing University of Technology Faculty of Information Science and Technology, Beijing, China., Beijing, China
Hongxiao Liu, Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.