Table 6.
Results of the proposed systems in this study for diagnosing pneumonia and tuberculosis and distinguishing between them.
| Techniques | Classes | Pneumonia | Tuberculosis | Normal | Accuracy % | |
|---|---|---|---|---|---|---|
| Hybrid method | VGG16 + SVM | 98.6 | 91.8 | 98.1 | 97.5 | |
| ResNet18 + SVM | 98.1 | 88.4 | 98.1 | 96.7 | ||
| Incorporating features before PCA | VGG16 + ResNet18 | 98.7 | 99.3 | 95.2 | 98.5 | |
| Incorporating features after PCA | VGG16 + ResNet18 | 96.5 | 99.4 | 92.8 | 97.8 | |
| Fusion features | ANN classifier | VGG16 and LDG | 100 | 99.9 | 97.6 | 99.6 |
| ResNet18 and LDG | 99.7 | 99.9 | 97.6 | 99.5 | ||