ConVnet BiLSTM for ASD Classification on EEG Brain Signal

As a neurodevelopmental disability, Autism Spectrum Disorder (ASD) is classified as spectrum disorder. The availability of an automated technology system to classify the ASD trait would have significant impact on paediatricians, it assist them in diagnosing children using quantifiable method. In this paper, we propose novel autism diagnosis method that based hybrid deep learning algorithms. This consists convolutional neural network (ConVnet) architecture merges two LSTM blocks (BiLSTM) with other direction propagation output state brain signal data from electroencephalogram (EEG) individuals; typically development (TD) and obtained Simon Foundation Research Initiative (SFARI) database state. For 70:30 distribution, accuracy 97.7 percent was achieved. Proposed methods outperformed current state-of-the art terms classification efficiency potential make contribution neuroscience research, demonstrated by results.