Nguyen V, Iyengar S, Rasheed H, et al. Comparison of Deep Learning and Clinician Performance Detecting Referable Glaucoma from Fundus Photographs in a Safety Net Population. Ophthalmology Science (2025), doi: https://doi.org/10.1016/j.xops.2025.100751.

• Figure 1 Patient-level algorithm and independent clinician performance with full years of experience (left) and precision–recall curve (right) when using patient-level expert panel reference labels. AUPRC = area under the precision–recall curve; PR = precision–recall; ROC = receiver operating characteristic; Sn = sensitivity; Sp = specificity.

The code for data preprocessing and model training is linked: https://github.com/informatics-isi-edu/eye-ai-exec/blob/main/ notebooks/VGG19/VGG19_Diagnosis_Train.ipynb.