Table 4 Performance of Eye-AD on the ROAD-I dataset for various patch sizes (i.e., 3 × 3, 5 × 5, 7 × 7 and 9 × 9)
From: Early detection of dementia through retinal imaging and trustworthy AI
Methods | Accuracy | Precision | F1-score | Kappa | AUC |
|---|---|---|---|---|---|
3 × 3 | 0.8859 | 0.8835 | 0.8840 | 0.6948 | 0.9341 |
5 × 5 | 0.8885 | 0.8862 | 0.8867 | 0.7018 | 0.9355 |
7 × 7 | 0.8938 | 0.8915 | 0.8914 | 0.7130 | 0.9302 |
9 × 9 | 0.8872 | 0.8856 | 0.8823 | 0.6851 | 0.9123 |
