Table 8 Simulation results for k-fold cross validation (k = 5 fold).
From: Advancing skin cancer diagnosis with deep learning and attention mechanisms
Metric | Fold 1 | Fold 2 | Fold 3 | Fold 4 | Fold 5 | Mean |
|---|---|---|---|---|---|---|
Accuracy | 97.7% | 97.5% | 97.8% | 97.6% | 97.9% | 97.7% |
Precision | 98.2% | 98.0% | 98.3% | 98.1% | 98.4% | 98.2% |
Sensitivity | 98.5% | 98.3% | 98.6% | 98.4% | 98.7% | 98.5% |
Specificity | 97.2% | 97.1% | 97.3% | 97.5% | 97.6% | 97.3% |
F1-Score | 98.3% | 98.1% | 98.4% | 98.2% | 98.5% | 98.3% |
Dice Score | 98.8% | 98.6% | 98.9% | 98.7% | 99.0% | 98.8% |
IoU | 99.4% | 99.3% | 99.4% | 99.3% | 99.5% | 99.4% |
