Intelligent Diagnosis of Autism Spectrum Disorder: Harnessing Machine Learning for Enhanced Early Detection

Autism Spectrum disease (ASD), a neurodevelopmental disease, is frequently accompanied by sensory difficulties such as over or undersensitivity to noises, smells, or touch. Although ASD’s major cause is genetic defects, early discovery and therapy can help ameliorate the condition. In recent years, intelligent diagnosis based on machine learning has grown in favour as a supplement to standard clinical procedures, which may be time-consuming and costly. In a clinical setting, autism spectrum disorder is often diagnosed by licenced professionals utilising labour-intensive and expensive methods. Because of this, scientists working in the domains of applied behavioural science, psychology, and medicine have created screening instruments in recent years for pervasive developmental disorders like autism and autism spectrum disorder, like the Modified Checklist for Autism and Autism Spectrum Quotient. The methods applied give suitable and comparable accuracy, as reported by other authors, and the application of supervised learning techniques like XGBoost has shown promising results. Both in terms of accuracy and other specific metrics sensitivity, etc. The XGBoost particularly performed well, giving an accuracy of about 90%, which is better than most other published work.