ARTIFICIAL INTELLIGENCE-ENHANCED ANALYSIS OF RETINAL VASCULATURE IN AGE-RELATED MACULAR DEGENERATION

Abstract

Purpose: To investigate associations between quantitative vascular measurements derived from intravenous fluorescein angiography (IVFA) and baseline characteristics on optical coherence tomography (OCT) in neovascular age-related macular degeneration (nAMD) patients. Methods: The authors prospectively recruited patients with active choroidal neovascularization secondary to AMD over 50 years old, presenting to a single center in Toronto, Canada from 2017 to 2023. Ultra-widefield IVFA images were processed using the artificial intelligence RETICAD FAassist system to extract quantitative information on blood flow, perfusion, and blood–retinal–barrier (BRB) permeability. Associations between IVFA parameters with functional and anatomical outcomes were examined using univariable and multivariable regression models. Results: Eighty-one nAMD eyes and seven healthy control eyes were included. Compared with healthy controls, BRB permeability in the central and peripheral retina was significantly higher in nAMD patients (P < 0.001). On univariable analysis, BRB permeability measured centrally was significantly associated with central macular thickness (P = 0.035), whereas perfusion and blood flow measured centrally were significantly associated with macular volume (P = 0.043 and 0.037, respectively). On multivariable analysis, BRB permeability remained significantly associated with central macular thickness (P = 0.026). Conclusion: Central BRB permeability measured on IVFA was significantly associated with baseline central macular thickness in nAMD patients. Future work should longitudinally explore associations between IVFA parameters and clinical characteristics in diverse nAMD populations.