Analysis and Prediction of Emotional and Interaction Patterns in Autistic and non-Autistic Children with VGG and Inception Models

Abstract

A neurological and developmental ailment that impacts children's behavioral and intellectual development is known as autism spectrum disorder (ASD). It frequently results in recurrent habits, narrow interests, communication problems, and trouble interacting with others. ASD's severity and long-term repercussions can be reduced with an early diagnosis. Traditional diagnostic methods are subjective, time-consuming, and require specialized expertise, leading to delays in intervention. The paper addresses the challenge of accurately diagnosing Autism Spectrum Disorder (ASD) in children. The aim is to develop an automated system using machine learning to predict and analyze the autistic children. By collecting and analyzing behavioral, cognitive, and physiological data, the system will identify key features that distinguish ASD. Multiple machine learning models will be developed and evaluated for accuracy and reliability. The final outcome will be a user-friendly diagnostic tool to assist healthcare professionals in early and precise ASD detection, improving early intervention and developmental outcomes.