Prediction of BRCA gene mutation status in epithelial ovarian cancer by radiomics models based on 2D and 3D CT images

Table 3 Cutoff value prediction ability of the 2D, 3D, and 2D + 3D joint radiomics models

	AUC	Accuracy	Sensitivity	Specificity	Positive value	Negative value	Cut-off
2D Training set	0.81 (0.71–0.91)	0.75 (0.63–0.84)	0.68	0.83	0.86	0.65	− 0.43
2D validation set	0.78 (0.61–0.96)	0.73 (0.54–0.87)	0.61	0.91	0.91	0.61
3D training set	0.80 (0.70–0.90)	0.75 (0.63–0.84)	0.71	0.81	0.84	0.65	− 0.61
3D validation set	0.75 (0.55–0.92)	0.74 (0.55–0.88)	0.84	0.58	0.76	0.70
2D + 3D Training set	0.91 (0.84–0.97)	0.86 (0.77–0.93)	0.95	0.74	0.84	0.92	0.14
2D + 3D validation set	0.82 (0.67–0.98)	0.73 (0.54–0.87)	0.611	0.91	0.91	0.61

ISSN: 1471-2342