Project Summary Optical coherence tomography angiography (OCTA) is a non-invasive, 3-dimensional imaging method to visualize and quantify microvasculature throughout the retina. The proposed study evaluates the clinical utility of OCTA measurements compared to standard structural measurements of the optic nerve head (ONH), retinal nerve fiber layer (RNFL) and macula measured using the current clinical imaging standard, spectral domain optical coherence tomography (OCT) to predict and detect progression of disease in glaucoma suspects and glaucoma patients. Our research, and that of others, has shown that superficial retinal vessel density (proportion of measured area composed of blood vessels) in the ONH region and macula is less dense in open angle glaucoma (OAG) eyes than in healthy eyes. Moreover, diagnostic accuracy is improved with increasing disease severity. Research from our laboratory suggests that the diagnostic accuracy of vessel density is similar to that of OCT-measured RNFL thickness, and that vessel density is reduced in retinal regions associated with localized visual field (VF) defects. These cross-sectional results strongly suggest that OCTA measurements reflect damage to tissues relevant to the pathophysiology of OAG. In a longitudinal study, the mean rate of change in macula vessel density was shown to be significantly faster in OAG eyes than in glaucoma suspect or healthy eyes. Finally, recent results indicate that shallow machine learning analyses of combinations of OCTA measurements can improve classification of healthy and glaucoma eyes compared to standard instrument measurements as can deep learning analyses of OCTA enface images. The current study provides a unique opportunity to extend the longitudinal OCTA data that has already been collected from 480 eyes racially diverse individuals for up to 10 years to investigate vessel density change over time in healthy, glaucoma suspect and OAG eyes, and to compare it to other imaging modalities. The aims of this study are 1) to improve our understandingof the relative change over time of OCTA and OCT measurements in healthy and diseased eyes to identify true disease-related change more accurately and 2) to improve our understanding of the risk of developing glaucomatous progression using statistical and deep learning-based analyses of multi- modal OCTA and OCT measurements. The proposed studies will enhance our understanding of age and glaucoma-related change in vessel density and retinal tissue thickness allowing us to predict and detect change more accurately, thus possibly slowing the rate of progression and consequently reducing the risk of reduction of vision related quality of life, including blindness.

Narrative Optical coherence tomography angiography (OCTA) allows measurement of vessel density in the retina that provides novel information on retinal microvasculature associated with the development and worsening of glaucoma in a racially diverse cohort. We propose to improve detection and prediction of glaucomatous progression by disassociating age-related changes in OCTA measurements from glaucoma-related changes, by employing deep-learning analyses and by modifying predictive models by adding OCTA measurements. Enhanced detection of progression should lead to earlier and more effective treatment, thus reducing the rate of disease worsening and preventing loss of vision-related quality of life (including blindness).