Table 16 Classification results obtained when the models are run with OASIS dataset.
1 vs. 1 | Models | Accuracy | Precision | Recall | F1 Score | AUC | MCC |
|---|---|---|---|---|---|---|---|
(5 slices selected manually) | |||||||
CN/AD | EfficientNetB2 + FPN (Our Model) ViT + EfficientNetB2 + FPN (Our Model) | 0.9165 0.9571 | 0.9390 0.8921 | 0.9372 0.8829 | 0.9345 0.8835 | 0.9654 0.9693 | 0.9341 0.8828 |
(5 slices selected SSIM) | |||||||
CN/AD | EfficientNetB2 + FPN (Our Model) ViT + EfficientNetB2 + FPN (Our Model) | 0.9783 0.9824 | 0.9671 0.9592 | 0.9598 0.9525 | 0.9755 0.9630 | 0.9890 0.9864 | 0.9735 0.9654 |
