Detection of Autism Spectrum Disorder (ASD) using Machine Learning Techniques: A Review

Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder described as a set of
conditions identified by different challenges like speech, social skills, non-verbal communication,
and repetitive behaviors. ASD depends on the factor of gender. The similarities and weaknesses
in autistic children and adults are distinct. The irreversible loss is observed if ASD is not
detected at earlier stage. Hence there is a need for automated techniques for early and accurate
detection. There are many developments in current research in the field of biomarkers for risk
assessment, diagnosis and tracking of disease progression. Machine learning used in health care
has made enhancements in diagnosis by processing and analysis of the huge amount of data.
Present research work focuses on automated methods of identification to diagnose ASD
accurately. Fusion method is used to combine any number of instruments, allowing data from
various reliable sources to be fused, all within an objective framework that can be converted to
the desired metric. Preprocessing techniques can be streamlined to incorporate techniques for
data fusion to minimize ambiguity in feature evaluation. For carrying out the research, 153
autism controls and 157 typical subjects of sMRI and fMRI for each is selected from Autism
Brain Imaging Data Exchange (ABIDE). This paper presents the overview of recent studies in
the semi-or fully-automatic computer-aided diagnosis of ASD and compares the parameters
visualized as methods applied, classes considered, features used, criteria of assessment and
results obtained. This paper also reveals the classification between ASD and TC subjects
for sMRI and fMRI using the K-NN classifier for different feature sets. Using feature
optimization and fusion of sMRI and fMRI images, classification efficiency can be enhanced.
Keywords: Autism Spectrum Disorder (ASD), Automatic Classification, Feature Extraction,
Image Fusion, Machine Learning Techniques, Typical Controls (TC)