Table 5 Result of Region Based Feature Connected Layer (RB-FCL) output for RSNA dataset
From: Multi Region-Based Feature Connected Layer (RB-FCL) of deep learning models for bone age assessment
Regressor | InceptionV3-RB-FCL | DenseNet121-RB-FCL | InceptionResNetV2-RB-FCL | ||||||
|---|---|---|---|---|---|---|---|---|---|
MAE | RMSE | MAPE% | MAE | RMSE | MAPE% | MAE | RMSE | MAPE% | |
LR | 7.48 | 10.22 | 9.415 | 7.48 | 10.22 | 9.413 | 7.48 | 10.22 | 9.413 |
KNN-R | 7.3 | 10.1 | 8.809 | 7.3 | 10.1 | 8.809 | 7.3 | 10.1 | 8.808 |
SVR | 7.85 | 11.64 | 13.44 | 7.85 | 11.64 | 13.44 | 7.85 | 11.64 | 13.44 |
RF-R | 7.14 | 10.13 | 8.252 | 7.14 | 10.13 | 8.284 | 7.14 | 10.12 | 8.36 |
DT-R | 9.67 | 13.87 | 10.52 | 9.57 | 13.73 | 10.46 | 9.67 | 13.7 | 10.67 |
ADB-R | 11.1 | 14.68 | 12.63 | 11.1 | 14.7 | 12.68 | 11.1 | 14.72 | 12.65 |
GB-R | 7.67 | 10.63 | 9.088 | 7.57 | 10.49 | 9.365 | 7.76 | 10.67 | 9.461 |