Ретиномика как инструмент прогнозирования глаукомы

Glaucoma is the leading cause of irreversible blindness, posing a major threat to eye health globally.¹ By 2040, an estimated 112 million people are expected to be affected by glaucoma, placing a substantial economic strain on both individuals and health care systems.² Sight loss from glaucoma can often be prevented through early diagnosis and treatment. However, the disease's asymptomatic nature in its early stages often leaves many individuals unaware of its onset, leading to late diagnosis and irreversible damage to the optic nerve.^3,4 In this situation, exploration of novel and robust biomarkers may aid early detection of glaucoma and the prevention of its irreversible late-stage complications.

The application of artificial intelligence for automated segmentation and quantification of color fundus photography (CFP) and OCT has significantly minimized the time and effort required to analyze geometric features in large datasets.^5,6 Previous studies have identified associations between a limited range of retinal vasculature features and glaucoma in CFP-based studies.⁷, ⁸, ⁹ Specifically, reduced retinal vascular caliber and fractal dimension, as well as decreased vascular tortuosity and branching angle, have been linked to glaucomatous status. In parallel, OCT-based investigations have demonstrated corresponding structural alterations, including thinning of the retinal nerve fiber layer (RNFL), ganglion cell-inner plexiform layer (GCIPL), and lamina cribrosa.¹⁰, ¹¹, ¹²