Skip to main content

Table 1 Post-processing rule justification

From: A hybrid framework for glaucoma detection through federated machine learning and deep learning models

x/n Rule

No. of True Prediction based on Rule

Justification

1/3 to Predict Glaucoma Diagnosis

14

Too lenient, may yield many false positives.

2/3 to Predict Glaucoma Diagnosis

85

Strikes a balance b/w sensitivity and specificity. Captures a significant number of true cases (85).

3/3 to Predict Glaucoma Diagnosis

373

Very conservative, may miss many true cases.

2/3 + 3/3 to Predict Glaucoma Diagnosis (Post-processing Rule)

458

Combines the above rules, capturing even more true cases (458).