TABLE 3.
Performance of deep learning models in elastography image classification.
| Elastography | Models | Accuracy | Precision | Recall | F1-score |
|---|---|---|---|---|---|
| Shear-wave | ResNet-50 | 0.41 | 0.57 | 0.32 | 0.35 |
| GoogLeNet | 0.81 | 0.79 | 0.80 | 0.80 | |
| DenseNet-121 | 0.87 | 0.87 | 0.85 | 0.86 | |
| EfficientNet-B6 | 0.71 | 0.76 | 0.65 | 0.67 | |
| Strain | ResNet-50 | 0.52 | 0.52 | 0.49 | 0.62 |
| GoogLeNet | 0.79 | 0.79 | 0.79 | 0.79 | |
| DenseNet-121 | 0.52 | 0.50 | 0.48 | 0.65 | |
| EfficientNet-B6 | 0.51 | 0.49 | 0.49 | 0.58 | |
| Doppler | ResNet-50 | 0.58 | 0.56 | 0.51 | 0.50 |
| GoogLeNet | 0.76 | 0.76 | 0.77 | 0.75 | |
| DenseNet-121 | 0.64 | 0.62 | 0.63 | 0.59 | |
| EfficientNet-B6 | 0.56 | 0.49 | 0.56 | 0.50 |