Table 3 Average predictive performance of each CNN for benign multi-class classification model
From: Recognizing pathology of renal tumor from macroscopic cross-section image by deep learning
CNN models | mi-AUC | ma-Sn | ma-Sp | mi-Acc |
|---|---|---|---|---|
EfficientNet-B4 | 0.9705 | 0.8558 | 0.9688 | 0.8947 |
ResNet-18 | 0.9307 | 0.7724 | 0.9290 | 0.8421 |
VGG-16 | 0.9575 | 0.7724 | 0.9567 | 0.8421 |