Table 6 Data augmenting both models. 5-fold cross-validation yields EndoSLAM results
From: Multimodal Biomedical Image Segmentation using Multi-Dimensional U-Convolutional Neural Network
Type | F = 1 | F = 2 | F = 3 | F = 4 | F = 5 | Avg |
|---|---|---|---|---|---|---|
Without DA | ||||||
 Traditional U-Net | 74.11 | 72.67 | 73.89 | 75.22 | 76.07 | 74.392 |
 MDU-CNN | 82.25 | 79.43 | 83.16 | 81.02 | 85.08 | 82.188 |
With DA | ||||||
 Traditional U-Net | 81.76 | 77.91 | 81.32 | 78.11 | 79.36 | 79.692 |
 MDU-CNN | 85.25 | 84.03 | 88.12 | 84.02 | 85.08 | 85.3 |