Autism spectrum disorder prediction system using machine learning and deep learning

Autism spectrum disease (ASD) is a neuro developmental illness that is both complicated and degenerative. A majority of known approaches use autism detection observation schedule (ADOS), pattern recognition, etc. to detect ASD with a small dataset, resulting in high accuracy but low generality. In this research, an ASD detection hybrid model is presented which is works on two different types of datasets. Firstly, behavioural datasets which works on logistic regression technique and secondly, facial dataset which works on based on convolutional neural network (CNN) Classifier in order to predict whether a person is suffering from autism or not. The suggested hybrid model which works on behavioural and facial dataset of a person beats state-of-the-art approaches in terms of accuracy, according to simulation findings. The proposed hybrid model had an average accuracy of 88% for the logistic regression model while it achieved an accuracy of 82.76% for the CNN model.