Implementation and Analysis of Various Machine Learning Algorithms for the Diagnosis of Autism Spectrum Disorder Using Matlab

Abstract

Early infancy may see the onset of symptoms, but it may take several visits to a pediatrician before a diagnosis is made. Long questionnaires, skilled medical experts, and occupational therapists are needed for subjective diagnosis of Autism spectrum disorder (ASD), which is a combination of developmental defects that causes social and behavioral deficits. This proposed work focuses on analyzing multiple machine learning models, constructed using MATLAB, to identify ASD using the ABIDE (Autism brain imaging data exchange) dataset, such as random forest, k nearest neighbors (K-NN), support vector machine (SVM), and decision trees. A considerable number of ASD patients and non-ASD controls were collected from 17 research and clinical institutes throughout the world as part of the global cooperative project ABIDE to create the dataset known as ABIDE. The main benefit of this mechanism is that it may quickly and objectively replace time-consuming exams now utilized in practice by utilizing recent advancements in machine learning and neuroimaging techniques.