Table 21 Ablation study performance with Raw images.
From: Feature fusion context attention gate UNet for detection of polycystic ovary syndrome
Model | With FCE images (%) | |||
|---|---|---|---|---|
Accuracy | Precision | Recall | F1-score | |
Baseline U-Net | 78.64 | 79.1 | 78.0 | 78.5 |
U-Net without FFCM | 79.10 | 79.5 | 78.6 | 79.0 |
U-Net with FFCM | 80.25 | 81.0 | 80.0 | 80.4 |
U-Net with Default Attention Gate | 81.36 | 82.0 | 81.2 | 81.3 |
U-Net with Modified Attention Gate | 82.10 | 82.6 | 81.9 | 82.2 |
FCAU-Net without FFCM and Attention Gate | 83.02 | 83.6 | 82.8 | 83.1 |
FCAU-Net without FFCM and Default Attention Gate | 85.12 | 85.8 | 84.7 | 85.2 |
FCAU-Net without FFCM and Modified Attention Gate | 86.27 | 87.0 | 85.9 | 86.4 |
FCAU-Net with FFCM and without Default Attention Gate | 88.15 | 88.7 | 87.6 | 88.1 |
FCAU-Net with FFCM and without Modified Attention Gate | 89.10 | 89.6 | 88.8 | 89.2 |
Proposed FCAU-Net | 90.51 | 91.2 | 90.0 | 90.6 |
