Detecting Autism Spectrum Disorder(ASD) Using CNN and its Methods in Deep Learning Techniques

Abstract

The world has seen the advent of numerous illnesses that cannot be medically recognized, such as Autism Spectrum Disorder (ASD). Autism, now called autism spectrum disorder (ASD), is a neuro developmental disorder. ASD is a developmental disability caused by differences in a person’s brain. People with ASD may behave, interact and learn in ways that are different from other people. They may have trouble with social interactions and with interpreting and using nonverbal and verbal communication. Autism is known as a “spectrum” disorder because there is wide variation in the type and severity of symptoms people experience. People of all genders, races, ethnicities, and economic backgrounds can be diagnosed with ASD. Like all people on the autism spectrum, people don’t naturally read social cues and might find it difficult to make friends. They can get so stressed by a social situation that they shut down. They don’t make much eye contact or small talk. People on the spectrum who are high-functioning can also be very devoted to routine and order. They might have repetitive and restrictive habits that seem odd to others. Early diagnosis based on different health and physiological characteristics seems feasible with the rising usage of machine learning-based models in predicting many human diseases. The proposed study with aims to use deep learning techniques like CNN, VGG16, VGG19, Densenet, Alexnet to predict the likelihood of ASD with a better degree of precision and minimal error rate and better accuracy. Further study is also required to find out much more optimized technique for the ASD detection