A Novel Approach to Personalized Prediction of Progression of Age-Related Macular Degeneration

The majority of patients with advanced AMD have severe vision loss. Despite the development of artificial intelligence algorithms for personalized AMD progression, we are still far from implementing tailored follow-up care and treatments for AMD patients. Our hypothesis is that the probability of AMD progression can be predicted by integrating imaging, genetic and clinical data in statistical predictive models, thereby improving personalized care.

Grantee institution at the time of this grant: University of Illinois Urbana-Champaign

The goal of our study is to predict the progression of age-related macular degeneration (AMD). We will first expand and validate our fully-automated image processing algorithms in predicting future choroidal neovascularization and geographic atrophy events to include genetic data and real-world longitudinal patient data. In our second aim, we will initiate a randomized, controlled clinical trial to test the viability of applying personalized AMD prediction models. This study integrates imaging, genetic, demographic, and clinical patient parameters from several data sources, with an existing pipeline for image feature extraction. This innovative approach will provide time-dependent probabilities for predicting advanced AMD. Testing algorithms in a clinical trial setting is another novel aspect of this research.

Our work will advance the AMD field by improving the identification of high-risk patients as candidates for more frequent screening and earlier treatment, leading to better clinical outcomes. Results may lead to developing a tool to predict the chances of AMD progression on a personalized, patient-by-patient basis.