Objective Explore the value of texture analysis of radiomics based on STIR to evaluate the activity in early AS.Methods A total of 43 patients with early AS were collected in our hospital,including 30 males and 13 females.Ac-cording to the ratio of 7∶ 3,all the patients were randomly divided into the training cohort(n =116)and verification cohort(n =56).Optimal feature subsets were obtained by the Mann-Whitney U test,the minimum-Redundancy Maximum-Rele-vancy(mRMR),and then the least absolute shrinkage and selection operator(LASSO)using these texture feature parame-ters.The optimal feature subset was used to construct the final prediction model,and Radscore was obtained by summing the selected features after being weighted by their coefficients.The effectiveness of Radscore value,SPARCC score and ADC value was evaluated by ROC curve.The correlations between Radscore values and the values of SPARCC score,ADC,BASDAI,ESR,CRP,ASDAS-ESR,ASDAS-CRP were analyzed by the Spearman correlation test.Results Eight tex-ture feature subsets were selected to identify the activity in early AS.In the training set/validation set,the best critical val-ue(AUC,AC,SEN,SPE,PPV,NPV)to distinguish the active group and the stable group of sacroiliac joint bone marrow e-dema in the early stage of AS:Radscore =51.46(0.81/0.87,76.72% /82.14% ,81.25% /90.65% ,66.67% /70.83% ,84.42% /80.56% ,61.54% /85.00% ).ROC curve efficiency:SPARCC score>Radscore value>ADC value,and Low correlation with BASDAI,ESR,CRP,ASDAS-ESR,ASDAS-CRP.Conclusion Radiomics analysis based on STIR texture analysis has a good prediction,and more objective evaluation method for AS activity.
Short time inversion recoveryRadiomicsAnkylosing spondylitisActivity
NETL
NSTL
万方数据
