Early Detection of Autism Spectrum Disorder (ASD) in Children using Machine Learning

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition that affects communication, social interaction, and behavior, necessitating early diagnosis for effective intervention. This study aims to improve the accuracy and efficiency of ASD screening for toddlers through the application of machine learning models utilizing the Q-CHAT-10 dataset. Following the CRISP-DM methodology, we conducted comprehensive data preparation, feature selection, and model evaluation. We compared the performance of three machine learning models: Logistic Regression, Decision Tree, and Artificial Neural Network (ANN). The ANN exhibited the highest performance, achieving an accuracy of 98.5% and an F1-Score of 98.5%, followed closely by the Decision Tree model with an accuracy of 98.23%. Logistic Regression, although less precise, maintained a reliable performance with an F1-Score of 91.02%. This research highlights the potential of AI-driven pre-diagnostic tools to expedite ASD screening processes, significantly reducing waiting times for assessments. Future work will focus on integrating clinical datasets and exploring multi-modal data, including eye-tracking and behavioral video analysis, to further enhance diagnostic accuracy and support early intervention strategies in real-world settings.