Autism Spectrum Disorder Detection Using Machine Learning Approach

Abstract

It is always a complex procedure to diagnosis autism spectrum disorder (ASD) because there is no particular medical test for autism, like a blood test, to make a diagnosis for the disorder. Autism spectrum disorder is defined by the disability and impairments of participating in social communication or the presence of restricted or repetitive behaviors, or both. It is a behaviorally diagnosed condition. To make a diagnosis, doctors look at the child's developmental history and behavior. Apparently, most children do not attain a proper diagnosis for autism until it is too late. In some cases, parents are reluctant to accept that their child's mental growth is not developing along with the child's physical growth. This lateness in diagnosis hinders a child's ability to get the help they need to keep developing. It is important to diagnose ASD as early as possible through monitoring, screening, and evaluating a child's development so that we can ensure proper care and support for an autistic child to help them reach their full potential. So, we are developing a system that will have the ability to diagnosis Autism and come up with a reliable and effective conclusion even without the help of a professional. We hope this system will be very helpful for those concerned parents who are worried about their child's growth and activities, at the time same time it will be very useful for the professionals.