Table 2.

Deep learning algorithm performance

Performance metric	Internal	External
AUC (95% CI)	0.93 (0.86 to 0.94)	0.81 (0.71 to 0.84)
Accuracy, %	89.5	79.3
Sensitivity	0.88	0.83
Specificity	0.87	0.67
PPV	0.71	0.55
NPV	0.95	0.90
Kappa*	0.69	0.48
Youden index	0.75	0.50
F1 scoreǂ	0.77	0.63

*Measures the agreement between the CNN model prediction and the clinical diagnosis. ^ǂMeasures the accuracy of the CNN model