Table 6 Performance metrics across 5 folds.
Fold | Accuracy | Precision | Recall | F1-score |
|---|---|---|---|---|
(a) Standard PCSA-ResNet integration | ||||
1 | 99.8% | 99.7% | 99.8% | 99.7% |
2 | 100% | 100% | 100% | 100% |
3 | 99.9% | 99.9% | 100% | 99.9% |
4 | 100% | 100% | 100% | 100% |
5 | 99.9% | 99.9% | 99.8% | 99.9% |
Mean | 99.92% | 99.9% | 99.92% | 99.9% |
(b) Pre-PCSA integration | ||||
1 | 96.3% | 95.8% | 96.2% | 96.0% |
2 | 96.5% | 96.2% | 96.5% | 96.3% |
3 | 95.9% | 95.5% | 96.0% | 95.7% |
4 | 96.4% | 96.0% | 96.2% | 96.1% |
5 | 96.1% | 95.9% | 96.3% | 96.0% |
Mean | 96.24% | 95.88% | 96.24% | 96.02% |
(c) Post-PCSA Integration | ||||
1 | 97.1% | 97.0% | 96.8% | 96.9% |
2 | 96.8% | 96.7% | 96.5% | 96.6% |
3 | 96.9% | 96.9% | 97.0% | 96.9% |
4 | 97.2% | 97.1% | 97.2% | 97.1% |
5 | 96.7% | 96.6% | 96.8% | 96.7% |
Mean | 96.94% | 96.86% | 96.86% | 96.84% |
