Insights into Imaging

Table 3 Comparison of the prediction performance of different models between the training and testing groups

From: Preoperative prediction of Lauren classification in gastric cancer: a radiomics model based on dual-energy CT iodine map

Models	Training set (n = 168)				Testing set (n = 72)
Models	AUC (95% CI)	Specificity	Sensitivity	Accuracy	AUC (95% CI)	Specificity	Sensitivity	Accuracy
Clinical model	0.611 (0.521–0.700)	0.785	0.473	0.565	0.630 (0.480–0.779)	0.708	0.437	0.527
R-IM model	0.756 (0.670–0.841)	0.642	0.830	0.767	0.730 (0.604–0.850)	0.458	0.791	0.680
R-MIX model	0.750 (0.673–0.826)	0.660	0.696	0.684	0.689 (0.566–0.811)	0.625	0.645	0.638
R-COMB model	0.855 (0.795–0.914)	0.714	0.848	0.803	0.803 (0.690–0.915)	0.750	0.770	0.763

Back to article page