Glaucoma is the second leading cause of blindness globally, with approximately 79.6 million people expected to be affected by the disease by 2020. Automated visual field (VF) testing remains the gold standard for identifying patients with glaucoma and judging worsening of disease. Approximately 5% of patients with glaucoma undergo rapid worsening of their VF test.

Early identification of these rapid progressors may allow ophthalmologists to perform medical and surgical intervention earlier to preserve visual function. The objective of this project is to develop machine learning algorithms that allow for sensitive and specific identification of patients at risk for undergoing rapid progression. This will be achieved with a two-step process:

1. We propose to expand a longitudinal dataset of 10,000+ VF (patient functional data) to include optic nervev imaging data routinely acquired in assessment of glaucoma and clinical data.
2. We will train machine learning models to predict the risk for future rapid worsening of NFs based on initial function (VF), structural (OCT) and clinical data.

We hypothesize that our machine learning models will be able to predict risk of future progression with high sensitivity and specificity.

Co PI: Jithin Yohannan, MD, MPH, Assistant Professor of Ophthalmology, Johns Hopkins Wilmer Eye Institute