ASD detection using an advanced deep neural network

Abstract

Autism Spectrum Disorder (ASD) is a neurological disorder which at present has become one of the most severe developmental disabilities causing social and behavioral changes in individuals. During the first 6 to 18 months of a person’s life, early indicators of ASD can be seen as further regression in development with ageing up to 36 months. Early recognition of the disorder is one of the solutions to the problem so that precautionary measures can be adopted against the disorder. In this proposed work, along with all categories, major emphasis is given to the unbalanced toddler data set. The original data sets are first, pre-processed following splitting of the pre-processed data into training and test data. For classification, a deep network model is implemented which is trained by the training data. The trained model then got tested by the test data for validating the performance of the classifier model to detect ASD class.

Subject Classification:

• Primary 91F99
• Secondary 65Z05

Keywords:

• ASD
• Standardization
• Feature extraction
• Classification
• Performance parameters