Unveiling Early Autism Spectrum Disorder Detection: A Comprehensive Study of Machine Learning and Deep Learning Approaches

According to research by the world health organization (WHO), approximately 0.63% of children are diagnosed with autism spectrum disorder (ASD). ASD commonly emerges during childhood and persists through adolescence and adulthood. Early detection of ASD is crucial as it can lead to effective interventions and improved outcomes. Machine learning and deep learning techniques have gained significant recognition for their efficacy in various fields including biomedicine and medicine. These models offer valuable tools for data interpretation in clinical diagnostics and decision-making processes. Consequently, researchers emphasize the utility of both machine learning and deep learning methods for disease screening. This study aims to leverage both machine learning and deep learning techniques to classify ASD cases using a dataset sourced from the UCI repository. The dataset encompasses pediatric cases. The classification task employed k-nearest neighbors’ algorithm (k-NN), support vector machine (SVM), and convolutional neural network (CNN). Among these, CNN outperformed SVM and k-NN, achieving a remarkable accuracy of 99.71%.