Table 5.
Experimental results for the adult datasets, trained exclusively on the adult dental dataset (1500 training images, 276 test images).
| Recall | Specificity | ACC | IOU | Dice | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| mean | std | mean | std | mean | std | mean | std | mean | std | |
| U-Net22 | 0.9459 | 0.0231 | 0.9795 | 0.0119 | 0.9719 | 0.0083 | 0.8858 | 0.0274 | 0.9392 | 0.0162 |
| R2 U-Net23 | 0.9351 | 0.0268 | 0.9838 | 0.0112 | 0.9724 | 0.0079 | 0.8892 | 0.0264 | 0.9411 | 0.0156 |
| PSPNet24 | 0.9164 | 0.0175 | 0.9827 | 0.0044 | 0.9670 | 0.0063 | 0.8693 | 0.0201 | 0.9299 | 0.0122 |
| Deeplab V3+25 | 0.9465 | 0.0218 | 0.9721 | 0.0130 | 0.9665 | 0.0088 | 0.8639 | 0.0284 | 0.9267 | 0.0171 |
Evaluation metrics include Recall, Specificity, Accuracy, Intersection over Union (IOU), and Dice index, with mean and standard deviation (std) values provided.