Table 7 Statistical results of the Densenet-201 model trained on normal and enhanced images.
From: Automated detection of polymicrogyria in pediatric patients using deep learning
Metric | Normal Model | Enhanced Model | Difference between enhanced and normal | Confidence Interval — Normal | Confidence Interval — Enhanced |
|---|---|---|---|---|---|
Accuracy | 0.9723 | 0.9939 | 0.0216 | 0.9646 – 0.9790 | 0.9900 – 0.9972 |
Precision | 0.9828 | 0.9989 | 0.0161 | — | — |
Recall | 0.9608 | 0.9888 | 0.028 | — | — |
F1-Score | 0.9717 | 0.9938 | 0.0221 | — | — |
AUC | 0.9972 | 0.9998 | 0.0027 | 0.9958 – 0.9984 | 0.9997 – 1.0000 |
Loss | 0.093 | 0.0233 | –0.0698 | — | — |
