Support vector machine based on clinical risk factors and CT radiomics for diagnosing of axial spondyloarthritis
MIAO Shouliang1, LIN Tingting1, XIAO Qinqin1, CHEN Dan2, YE Lusi2, ZHENG Xiangwu1.
1.Department of Radiology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou 325015,China; 2.Department of Rheumatology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou325015, China
MIAO Shouliang,LIN Tingting,XIAO Qinqin, et al. Support vector machine based on clinical risk factors and CT radiomics for diagnosing of axial spondyloarthritis[J]. JOURNAL OF WEZHOU MEDICAL UNIVERSITY, 2022, 52(3): 199-204.
Abstract：Objective: To construct a prediction model of axial spondyloarthritis (axSpA) based on CT imaging and clinical features by support vector machine (SVM). Methods: A group of 568 patients diagnosed as axSpA (n=319) or non-axSpA (n=249) between October 2012 and February 2019 were included retrospectively,who were randomly divided into training and validation cohorts at a ratio of 7:3. The volume of interest (VOI) was manually sketched on the sacroiliac joint CT and the radiomics were extracted. The mRMR and LASSO algorithms were used to select the optimal parameters. Univariate and multivariate Logistic regression analysis was made to find the clinical risk factors of axSpA. Finally, SVM was used to build clinical, radiomics and clinicalradiomics combined models, and their performance was evaluated by the receiver operating characteristic curve (ROC) and Delong test. Results: The clinical-radiomics model showed the best diagnostic efficacy in the validation cohort and higher area under curve (AUC) values (AUC=0.91) than radiomics model (AUC=0.83) or clinicalmodel (AUC=0.81) (P<0.05), with high accuracy (0.83), sensitivity (85.2%) and specificity (79.7%). Conclusion:Support vector machine based on CT radiomics and clinical risk factors has high value in the diagnosis of axSpA.
