Table 6 95% Confidence Intervals for performance metrics of the EfficientNetB4 model used for classification of Plus disease on the validation set (960 images).
From: Automated diagnosis of plus form and early stages of ROP using deep learning models
Metric | Point estimate | 95% CI lower | 95% CI upper |
|---|---|---|---|
Accuracy | 0.996 | 0.9942 | 1.0000 |
F1 Score | 0.98 | 0.9786 | 1.0000 |
Sensitivity | 0.99 | 0.9846 | 1.0000 |
Specificity | 0.997 | 0.9958 | 1.0000 |