CT-based radiomics signature of visceral adipose tissue and bowel lesions for identifying patients with Crohn’s disease resistant to infliximab

Insights into Imaging

Table 3 Predictive performance of radiomics models based on different features in differentiating PNR from PR in the training and validation cohorts

Variable	Accuracy	Sensitivity	Specificity	AUC (95% CI)	p
Training cohort (PNR/PR = 33/79)
VAT radiomics model	0.705	0.727	0.696	0.761 (0.672–0.837)	< 0.001
Bowel radiomics model	0.732	0.848	0.684	0.832 (0.750–0.896)	< 0.001
VAT-bowel radiomics model	0.821	0.909	0.785	0.873 (0.797–0.928)	< 0.001
Internal validation cohort (PNR/PR = 14/34)
VAT radiomics model	0.688	0.571	0.735	0.737 (0.590–0.854)	0.001
Bowel radiomics model	0.708	0.857	0.647	0.784 (0.641–0.889)	< 0.001
VAT-bowel radiomics model	0.813	0.714	0.853	0.840 (0.706–0.930)	< 0.001
External validation cohort (PNR/PR = 22/49)
VAT radiomics model	0.662	0.727	0.633	0.714 (0.595–0.815)	0.001
Bowel radiomics model	0.690	0.590	0.735	0.799 (0.687–0.885)	< 0.001
VAT-bowel model	0.817	0.636	0.898	0.833 (0.726–0.911)	< 0.001

Accuracy, sensitivity, and specificity of the radiomic model in training and validation cohorts were calculated with the cut-off value of 0.280 (VAT radiomics model), 0.190 (bowel radiomics model), and 0.268 (VAT-bowel radiomics model), respectively, which maximizes the Youden index in the training cohort. p value is the significance level of comparison of AUC with that of random case (AUC = 0.5). PNR, primary nonresponse to infliximab therapy; PR, response to infliximab therapy. AUC, area under ROC curve; CI, confidence interval. VAT radiomics model, radiomics model based on the features extracted from visceral adipose tissue; bowel radiomics model, radiomics model based on the features extracted from the whole inflamed bowel; VAT-bowel radiomics model, a combination of the VAT and bowel radiomics models