Table 14 Comparative analysis of the diagnostic accuracy of the hybrid deep learning algorithms.
Model | Accuracy (LB, UB) | Sensitivity (LB, UB) | Specificity (LB, UB) | Precision (LB, UB) | Recall (LB, UB) | AUC (LB, UB) |
|---|---|---|---|---|---|---|
CNN | (82,87) | (79,86) | (80,85) | (81,87) | (82,86) | (0.82,0.87) |
CNN-AlexNet | (84,89) | (81,85) | (84,86) | (83,88) | (82,89) | (0.84,0.89) |
CNN-VGG | (83,90) | (80,91) | (81,89) | (80,91) | (82,92) | (0.83,0.90) |
CNN-NiN | (85,91) | (82,90) | (83,90) | (83,91) | (84,90) | (0.85,0.91) |
CNN-GoogleNet | (83,91) | (81,90) | (82,89) | (81,90) | (81,91) | (0.83,0.91) |
CNN-ResNet | (84,93) | (82,90) | (80,92) | (81,92) | (83,93) | (0.84,0.93) |
CNN-DenseNet | (82,92) | (85,93) | (84,90) | (86,90) | (83,91) | (0.82,0.92) |
CNN-VGG + SGD | (88,96) | (86,95) | (85,96) | (86,95) | (88,96) | (0.88,0.96) |
CNN-GoogleNet + SGD | (89,96) | (88,95) | (87,95) | (87,95) | (88,96) | (0.89,0.96) |
CNN-DenseNet + SGD | (87,96) | (85,95) | (86,95) | (88,96) | (87,95) | (0.87,0.96) |
