Table 8 Ablation Study Results.
From: Automated detection of polymicrogyria in pediatric patients using deep learning
Pipeline | Test Loss | Test Accuracy | Test Precision | Test Recall | Test F1 |
|---|---|---|---|---|---|
Original Images | 0.3759 | 0.8080 | 0.7249 | 0.9922 | 0.8378 |
Grayscale Conversion | 0.3759 | 0.8080 | 0.7249 | 0.9922 | 0.8378 |
Grayscale + Normalization | 0.3683 | 0.8168 | 0.7344 | 0.9922 | 0.8441 |
Grayscale + Normalization + CLAHE | 0.2768 | 0.8932 | 0.8293 | 0.9900 | 0.9026 |
Grayscale + Normalization + CLAHE + Bilateral Filtering | 0.1229 | 0.9696 | 0.9680 | 0.9712 | 0.9696 |
Full Pipeline (Includes Edges Blend) | 0.1183 | 0.9696 | 0.9569 | 0.9834 | 0.9700 |
