Table 2 Performance metrics evaluated from the segmentation models used for blood vessel segmentation on Validation set.
From: Automated diagnosis of plus form and early stages of ROP using deep learning models
Model | Accuracy | Recall | IoU | Dice |
|---|---|---|---|---|
DeepLabV3 | 0.9036 | 0.6228 | 0.4254 | 0.5967 |
DeepLabV3+ | 0.9451 | 0.7172 | 0.5315 | 0.6932 |
Unet | 0.91 | 0.6621 | 0.4564 | 0.6381 |
MAnet | 0.9749 | 0.8358 | 0.7453 | 0.8534 |
Unet++ | 0.9766 | 0.8385 | 0.7562 | 0.8584 |
