Prediction of Autism Spectrum Disorder Using Efficient Net
Mohana sree venkata sai krishna Narala, Sandeep Vemuri, Chandrika Kattula
Abstract
Autism Spectrum Disorder is a neurodevelopmental condition that can have a lasting impact on a person's ability to communicate, learn language, interact socially, and process information.Early detection of this disorder in children helps to improve their intellectual ability. This model helps to detect autism spectrum disorder. So, to predict whether the person is either autistic or a typically developing child is done using facial images. Our findings facilitate the distinction between children diagnosed with Autism Spectrum Disorder and those who are typically developing by analyzing their facial characteristics. Autism Image data dataset is used which consists of 2530 facial images in training set, 300 facial images are present in the test set which are used for evolution of model. The Efficient Net convolutional neural network was utilized to build this model, which achieved a accuracy level of 88.