ASD Classification in Adolescent and Adult Utilizing Deep Neural Network

Autism Spectrum Disorder (ASD) is one of the neurological illnesses affecting the behaviour and communicative skills of an individual. It hampers the recognition capability of an individual. Hence it is the primary responsibility towards the affected individuals with ASD for early detection to minimize its effect. ASD clinical diagnosis procedure is lengthy and expensive. So, against the procedure, ASD datasets are stored in authenticated sites like Kaggle and UCI Machine Learning (ML) repository to carry out clinical research. The data from all the category of individuals including adult, adolescent, child and toddler got collected by a mobile based ASDTest app with certain screening questions. The proposed method covered the category of adolescent and adult datasets with implementation of Landmark Isomap for dimension reduction and then improved Deep Neural Network prediction with classification (iDNNPC) architecture for detecting ASD class. The evaluation of performance parameters confirmed the accomplishment of i-DNNPC classifier model.