Automatic Visual Impairment Detection System for Age-related Eye Diseases through Gaze Analysis

摘要

Visual impairment associated with Age-related Macular Degeneration (AMD) often results in a central scotoma which is an alteration in the central vision, leading to distortion or loss of vision. Current methods for assessing visual performance such as Amsler grid and Microperimetry are typically manual and have limitations as an indicator of visual field. In this paper, we present an automated system for detecting visual impairment through gaze tracking (AVIGA). Two types of assessments namely, Impulse Stimuli Response (ISR) test and Pursuit Stimuli Response (PSR) test were implemented in AVIGA system. A Support Vector Regression (SVR)-based approach is applied on the assessment results to differentiate the severity of visual impairment. The results show that AVIGA system is well-correlated to visual acuity test (VA) and performs better in identifying presence of visual impairments in eyes, compared to Microperimetry.