Detection of Brain Disorders Using Ocular Responses

This paper introduces an innovative approach to early detection of Autism Spectrum Disorder (ASD) by integrating Artificial Intelligence (AI) with eye-tracking data, addressing the challenge of delayed intervention, especially in resource-limited settings. Drawing from previous research and advancements, the work aims to present an AI-driven solution for early ASD detection using ocular responses. By tracking eye movements via webcam during video viewing, the process generates a graph representing eye scan paths, which are then transformed into images for classification to determine ASD presence. The methodology integrates a Convolutional Neural Network (CNN) model with eye-tracking data, facilitating early ASD detection. Experimental results highlight the approach's efficiency, demonstrating benefits in cost, labor, and space ef-ficiencies. In essence, this work offers an inventive solution for early detection of brain disorders, particularly ASD, with potential for future advancements in ASD detection methodologies. Future research may explore the applicability of ocular responses in diagnosing other brain disorders.