| H&E |
Hematoxylin and eosin |
| CAD |
Computer-aided diagnosis |
| CNN |
Convolutional neural network |
| xAI |
Explainable artificial intelligence |
| CAM |
Class activation mapping |
| Grad-CAM |
Gradient-weighted class activation mapping |
| LIME |
Local interpretable model-agnostic explanations |
| UCSB |
Breast cancer dataset |
| CR |
Colorectal cancer dataset |
| LG |
Liver tissue dataset |
| OED |
Oral epithelial dysplasia dataset |
| SGDM |
Stochastic gradient descent |
| F |
Fractal techniques |
| D |
DenseNet-121 |
| E |
EfficientNet-b2 |
| I |
Inception-V3 |
| R |
ResNet-50 |
| V |
VGG-19 |
|
Grad-CAM representation via DenseNet-121 |
|
Grad-CAM representation via EfficientNet-b2 |
|
Grad-CAM representation via ResNet-50 |
|
Grad-CAM representation via Inception-V3 |
|
Grad-CAM representation via VGG-19 |
|
LIME representation via DenseNet-121 |
|
LIME representation via EfficientNet-b2 |
|
LIME representation via Inception-V3 |
|
LIME representation via ResNet-50 |
|
LIME representation via VGG-19 |
| VP |
True positives |
| VN |
True negatives |
| FP |
False positives |
| FN |
False negatives |