PRELIMINARY STAGE AUTISM SPECTRUM DISORDER DETECTION

Abstract

Autism Spectrum Disorder (ASD) is a disorder related to brain development that impact the way a child perceives, learns, socializes and communicates with others. Although autism can be diagnosed at any age, it is said to be a developmental disorder because symptoms generally appear in the first two years of life and can persist till adulthood if not treated at same time. Child with autism suffers from limited behavior, interaction and communication. Diagnosing ASD can be difficult since there is no medical test, like a blood test, to diagnose the disorders. Specialists and doctors look at the child’s behavior and development for certain span of time to make a diagnosis. Previously, there were few methods for early detection like multiple screening tools, combined questionnaire and video capturing using Machine Learning Classifier. The proposed system will focus on early detection of ASD using Convolutional Neural Network (CNN). It intends to predict the level of the disorder among the infants at early stage so that they could get preliminary treatment. Taking images of the child as input, the system analyzes the image and determines the possibility of autism in the child.