Table 5 Predictive performance of each class for ResNet-18 architecture
From: Recognizing pathology of renal tumor from macroscopic cross-section image by deep learning
Metrics | pRCC | ccRCC | chRCC | OMT | AML | MCRN-LMP | RO | OBT |
|---|---|---|---|---|---|---|---|---|
AUC | 0.7791 | 0.8012 | 0.7656 | 0.85 | 0.9487 | 1.0 | 0.7292 | 0.8824 |
Sn | 0.4 | 0.8246 | 0.375 | 0.5 | 0.9231 | 1.0 | 0.6667 | 0.5 |
Sp | 0.8955 | 0.7333 | 0.875 | 1.0 | 0.8333 | 1.0 | 1.0 | 0.8824 |
Acc | 0.8611 | 0.8056 | 0.8194 | 0.9861 | 0.8947 | 1.0 | 0.9474 | 0.8421 |