Using apparent diffusion coefficient maps and radiomics to predict pathological grade in upper urinary tract urothelial carcinoma

Table 4 Performance of the mean ADC value and radiomic model

Model	AUC (95% CI)	Sensitivity (95% CI)	Specificity (95% CI)	Accuracy (95% CI)	P Value
Training set
Mean ADC value	0.698 (0.625, 0.772)	0.504 (0.416,0.592)	0.893 (0.778, 1.000)	0.576 (0.573,0.579)
Radiomics model	1.000 (1.000, 1.000)	1.000 (1.000, 1.000)	1.000 (1.000, 1.000)	1.000 (1.000, 1.000)	< 0.001^*
Test set
Mean ADC value	0.628 (0.474,0.782)	0.528 (0.394,0.663)	0.727 (0.464,0.990)	0.562 (0.555,0.570)
Radiomics model	0.786 (0.651, 0.921)	0.755 (0.639, 0.871)	0.727 (0.464, 0.990)	0.750 (0.744, 0.756)	0.071 ^*

Note ^*Comparison of ROC curve performance between the mean ADC value and the radiomics model using the DeLong test. AUC, area under curve; CI, confidence interval. ADC, Apparent diffusion coefficient

ISSN: 1471-2342