Brain Wave Based Autism Spectrum Disorder Detection -A Machine Learning Approach

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social communication and the presence of restricted interests and repetitive behaviors. Traditional diagnostic methods are often subjective, time-consuming, and reliant on specialized clinicians, leading to delays in diagnosis and treatment. This Paper leverages machine learning (ML) to develop an automated system for ASD detection using Electroencephalography (EEG) data, a noninvasive method for recording brain activity. Temporal and spatial feature extraction techniques, combined with statistical measures (mean, standard deviation, skewness, kurtosis) and frequency-based metrics (mean and median frequencies), were applied to EEG signals. These features were used to train a Logistic Regression model, which demonstrated promising accuracy in classifying EEG data as normal or abnormal, indicative of ASD-related patterns. The findings underscore the potential of ML-integrated EEG analysis to enhance early ASD diagnosis, improving accessibility, diagnostic speed, and accuracy. Further validation in clinical settings and addressing dataset diversity are essential to enable its real-world application.