Machine Learning Based Automatic Detection of Genetic Diseases in Pediatric Age Using Pupillometry

Abstract

Hereditary ocular disorders present significant visual impairment challenges in pediatric populations, resulting in substantial visual dysfunction and potential blindness during developmental years. These conditions encompass both peripheral and central retinal pathologies, with diagnostic complexity arising from extensive clinical and genetic heterogeneity involving more than 200 identified causative genes. Traditional diagnostic approaches rely on comprehensive clinical evaluations, including invasive procedures often unsuitable for infants and young children. This research proposes an innovative methodology utilizing Chromatic Pupillometry techniques for assessing retinal functionality across different layers. The study presents a groundbreaking Clinical Decision Support System incorporating Machine Learning algorithms with chromatic light stimulation protocols to enhance pediatric genetic disease detection capabilities.