Systematic review of the use of machine learning combined with radiomics in the diagnosis and differential diagnosis of tuberculous spondylitis
10.3969/j.issn.1002-1671.2025.08.020
- VernacularTitle:机器学习结合影像组学在结核性脊柱炎诊断与鉴别诊断的系统评价
- Author:
Yi CAI
1
;
Ruihan LI
1
;
Qian WU
1
;
Hui GUO
1
Author Information
1. 新疆医科大学第四附属医院放射科,新疆 乌鲁木齐 830000;新疆医科大学第四临床医学院,新疆 乌鲁木齐 830000
- Publication Type:Journal Article
- Keywords:
artificial intelligence;
radiomics;
machine learning;
tuberculous spondylitis
- From:
Journal of Practical Radiology
2025;41(8):1348-1351,1360
- CountryChina
- Language:Chinese
-
Abstract:
Objective To evaluate the efficacy of various machine learning models combined with radiomics in the diagnosis and differential diagnosis of tuberculous spondylitis.Methods A literature search was conducted across six databases(PubMed,Web of Science,Embase,Wanfang,VIP,and CNKI)to collect all relevant articles on the diagnosis and differential diagnosis of tuberculous spondylitis published from their inception to September 1,2024.Results Eight studies were included(5 in English and 3 in Chinese),with 4 on the differential diagnosis between tuberculous spondylitis and brucellar spondylitis,2 on pyogenic spondylitis,and 2 on spinal metastases.The various models used in the included studies demonstrated good diagnostic performance.Conclusion Artificial intelligence shows potential in assisting with the diagnosis and differential diagnosis of tuberculous spondylitis.However,its clinical application faces numerous limitations and risks.This study provides new insights into the diagnosis and differential diagnosis of tuberculous spondylitis and offers valuable experience and feasibility for future multicenter and prospective studies involving machine learning combined with radiomics.
