Spectral CT-based radiomics signature for distinguishing malignant pulmonary nodules from benign

BMC Cancer

Table 2 Comparison of AUCs between the radiomics model, conventional model, and integrated model

Model	Cut-off	Training dataset				Testing dataset
Model	Cut-off	AUC (95%CI)	SEN	SPE	ACC	AUC (95%CI)	SEN	SPE	ACC
Radiomics model	0.91	0.96 (0.925–0.996)	0.94	0.91	0.92	0.96 (0.914–0.996)	0.87	0.90	0.87
Conventional model	0.60	0.88 (0.823–0.941)	0.86	0.81	0.84	0.86 (0.767–0.948)	0.82	0.70	0.76
Integrated model	1.01	0.97 (0.940–0.997)	0.91	0.93	0.91	0.97 (0.928–1.000)	0.91	0.95	0.92

AUC Area under ROC curve, 95% CI 95% confidence interval, SEN Sensitivity, SPE Specificity, ACC Accuracy, Radiomics model The model combining optimal radiomics scores based on 65 keV images of AP and VP, Conventional model The model based on significant clinical characteristics and spectral quantitative parameters, Integrated model The model combining radiomics model and Z_eff-AP

ISSN: 1471-2407