Research Article
Multi-Features-Based Automated Breast Tumor Diagnosis Using Ultrasound Image and Support Vector Machine
Table 3
The classification result of using classical deep neural network.
| Pretrained models | Accuracy | Sensitivity | Specificity | F1-score |
| VGG16 | 0.84 | 0.86 | 0.82 | 0.86 | VGG19 | 0.81 | 0.81 | 0.77 | 0.82 | ResNet50 | 0.78 | 0.77 | 0.77 | 0.85 | InceptionV3 | 0.78 | 0.83 | 0.7 | 0.81 |
|
|