Table 5 The performance evaluation of applied ML techniques on features extracted using novel MobLG-Net
From: Novel transfer learning based bone fracture detection using radiographic images
Model | Accuracy | Target class | Precision | Recall | F1 Score | Support |
|---|---|---|---|---|---|---|
KNN | 0.98 | fractured | 0.97 | 0.99 | 0.98 | 892 |
non-fractured | 0.99 | 0.97 | 0.98 | 881 | ||
Average | 0.98 | 0.98 | 0.98 | 1773 | ||
LGB | 0.99 | fractured | 0.99 | 0.99 | 0.99 | 892 |
non-fractured | 0.99 | 0.99 | 0.99 | 881 | ||
Average | 0.99 | 0.99 | 0.99 | 1773 | ||
LR | 0.99 | fractured | 0.99 | 0.99 | 0.99 | 892 |
non-fractured | 0.99 | 0.99 | 0.99 | 881 | ||
Average | 0.99 | 0.99 | 0.99 | 1773 | ||
RF | 0.98 | fractured | 0.98 | 0.98 | 0.98 | 892 |
non-fractured | 0.98 | 0.98 | 0.98 | 881 | ||
Average | 0.98 | 0.98 | 0.98 | 1773 |