PROJECT SUMMARY
Glaucoma is a leading cause of irreversible blindness in Black, Hispanic, and Latino communities in the United
States. These populations often face substantial barriers in accessing timely and effective diagnostic services,
a challenge intensified by socioeconomic and systemic healthcare inequalities. Such obstacles frequently lead
to late-stage presentation, drastically increasing risk of blindness. This reality highlights the urgent need for the
development of affordable, accessible, and equitable screening approaches for glaucoma. Although imaging
technologies such as Spectral domain-optical coherence tomography (SDOCT) can provide highly reproducible
and accurate quantitative assessment of glaucomatous damage, their application in widespread screening is
unfeasible, given the high cost and operator requirements. Fundus photography is a low-cost alternative that
has been used successfully in teleophthalmology. However, subjective grading of fundus photos for glaucoma
is poorly reproducible and highly inaccurate. We propose a new paradigm for assessing glaucomatous damage
by training a deep learning (DL) convolutional neural network to provide objective quantitative estimates of neural
damage from fundus photographs. In our Machine-to-Machine (M2M) approach, a DL network analyzes fundus
photos to predict quantitative measurements of glaucomatous damage provided by SDOCT, such as retinal
nerve fiber layer (RNFL) thickness. Our preliminary results showed that the M2M predictions have very high
agreement and correlation with original SDOCT observations. Most importantly, the M2M was shown to detect
glaucoma, predict damage in suspects and track longitudinal change with a performance comparable to SDOCT.
In this proposal, we aim at refining and validating the M2M model for equitable opportunistic and community-
based glaucoma detection in underserved communities. In Aim 1, we will refine the M2M model using large
clinic- and population-based datasets applying innovative frameworks such as vision transformers and
generative AI to ensure equitable and effective glaucoma screening across diverse populations. In Aim 2, we
will implement the M2M model for opportunistic glaucoma screening at the Bascom Palmer Eye Institute (BPEI)
eye-specific emergency room (ER), the highest volume eye-specific ER in the nation, that serves a large
population from underserved communities and high-risk ethnic/racial groups for glaucoma. We hypothesize that
the M2M model will be feasible, equitable, and have high acceptance and accuracy rates for glaucoma
opportunistic screening. In Aim 3, we will apply the M2M model for detecting glaucoma in a community-based
screening program that targets underserved populations in several South Florida counties. We hypothesize that
the model will have high accuracy to detect glaucoma and will outperform clinician-based assessment. This
proposal harnesses the power of AI to bridge existing gaps in glaucoma care. If successful, it will validate an
equitable AI tool that could be seamlessly integrated into diverse healthcare settings, promoting equal access to
early and accurate glaucoma diagnosis.
