Table 1 Patch-wise model performances in the tumor classifier (upper) and the EBV classifier (lower).
From: Deep learning model to predict Epstein–Barr virus associated gastric cancer in histology
Model | Patch size (pixels) | Accuracy | NPVa | Specificity | Sensitivity | Precision | F1-score | |
|---|---|---|---|---|---|---|---|---|
Tumor classifier | InceptionV3 | 256 × 256 | 0.98 | 0.97 | 0.96 | 0.98 | 0.98 | 0.98 |
512 × 512 | 0.98 | 0.99 | 0.95 | 0.99 | 0.98 | 0.99 | ||
ResNet50 | 256 × 256 | 0.98 | 0.97 | 0.96 | 0.99 | 0.98 | 0.98 | |
512 × 512 | 0.98 | 0.98 | 0.97 | 0.99 | 0.99 | 0.99 | ||
EBV classifier | InceptionV3 | 256 × 256 | 0.93 | 0.99 | 0.93 | 0.90 | 0.56 | 0.69 |
512 × 512 | 0.99 | 0.99 | 0.99 | 0.92 | 0.92 | 0.92 | ||
ResNet50 | 256 × 256 | 0.92 | 0.97 | 0.94 | 0.71 | 0.54 | 0.61 | |
512 × 512 | 0.98 | 0.99 | 0.99 | 0.91 | 0.86 | 0.89 |
