Autism Spectrum Disorder Pre-Diagnosis

Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition affecting communication, behavior, and social interactions. Early detection is crucial for timely intervention, and this project introduces a machine learning-based approach to ASD diagnosis using behavioral and physiological data. Various preprocessing techniques were applied to enhance data quality, and multiple machine-learning models were implemented to identify ASD patterns. Feature selection methods improved model performance, achieving high accuracy in distinguishing ASD from neurotypical individuals. Comparative analysis highlights the efficiency of this approach, which contributes to medical diagnostics by automating ASD detection. Future enhancements may include expanding the dataset, integrating genetic and imaging data, and improving model interpretability to ensure broader clinical relevance. This research underscores the role of artificial intelligence in advancing healthcare and improving diagnostic processes.