Digital twin for ocular hemodynamics: Combining physiology-based modeling and machine learning for personalized glaucoma care.

Abstract

This work investigated the combined effects of intraocular pressure (IOP) and blood pressure (BP) on retinal hemodynamics and glaucoma progression using a novel, physiology-based digital twin for ocular hemodynamics (DT-OH). The DT-OH integrates a mathematical model of ocular physiology with machine learning to simulate retinal blood flow dynamics based on individualized IOP and BP inputs. The DT-OH was applied to clinical data from the Indianapolis Glaucoma Progression Study (IGPS) to characterize how IOP and BP jointly influence retinal hemodynamics and their association with glaucoma progression. The DT-OH identified three distinct hemodynamic profiles based on the combined effects of IOP and BP. Membership in one specific profile at baseline was associated with a significantly higher risk of glaucoma progression. These profiles reflect distinct patterns of ocular blood flow regulation and provide physiological insight into the interplay between systemic and ocular factors in glaucoma. These findings enhance our understanding of glaucoma pathophysiology and support the development of personalized risk assessment tools that account for both IOP and BP.