Machine Learning for Autism Spectrum Disorder Detection: A Systematic Survey

A neurodevelopmental illness called autism spectrum disorder (ASD), often referred to as autism causes issueswith social communication. These issues are evident in the way ASD youngsters talk, according to prior study. This shows that quantitative speech variables may be used to judge the severity of ASD.A common way research is one method of identifying ASD behavioral child's expression of signals. Using Machine Learning can obtain the data of children having the autism spectrum disorder. Through last years, applying magnetic resonance imaging (MRI) to detect common patterns type community is a set of stacked auto encoders trained in an unsupervised manner and multilayer perception trained in a supervised manner. Quantitative assessment was completed on 817 multi-site Autism Brain Imaging I Global Data Exchange I (ABIDE I) data cases, including 368 ASD patients and 449 control subjects, and received an accuracy of type 85% when assessed using a set of classifiers.