Autism Spectrum Disorder Detection Using MobileNet

Abstract

Autism Spectrum Illness (ASD), an evolution of the brain disorder, is commonly related with sensory difficulties, such as excessive or insufficient sensitivity to sounds, scents, or touch. Autism Spectrum Disorder (ASD) is evolving at a faster rate than ever before. By screening tests autism detection is very expensive and time consuming. With the advancement of Deep Learning (DL),autism can be predicted from a young age. In this paper we are using Convolutional Neural Network (CNN) with Transfer Learning (TL) models to classify the disease and we will suggest the precautions if it is detected as autism. Here we consider the Autism Master Dataset (AMD) from kaggle.com website, which contains two classes (Autism, Non_Autism). By using this models, we are obtaining good accuracy.