DEEP LEARNING MODELS FOR ANALYZING PUPILLARY DYNAMICS IN GENETIC DISORDER DETECTION

Abstract

The primary objective of this initiative is to develop deep learning models that will investigate pupillary dynamics in the diagnosis of genetic diseases. Variations in pupil size, reflex speed, and light adaptation patterns, among other pupillary responses, are significant neurological and physiological markers that may be associated with underlying genetic disorders. The objective of this research is to analyze and comprehend pupillary behavior as captured by eye-tracking and imaging sensors by employing cutting-edge deep learning techniques, including CNNs, RNNs, and LSTM models. The proposed method seeks to improve the accuracy, efficiency, and early diagnosis of genetic abnormalities in comparison to current clinical assessment procedures by extracting complex spatial and temporal information from pupillary movement data. A non-invasive, cost-effective, and reliable diagnostic framework that integrates AI with neurological and ocular indicators can enable healthcare providers to benefit from predictive genetic analysis and individualized treatment plans.