No abstract

No abstract

No abstract

No abstract

Autism spectrum disorder is a developmental disorder that has affected many children around the globe in recent years. It is possible to reduce the severity of the symptoms when the affected children are identified and treated early. Hence, early detection and treatment of this neurodevelopmental disorder significantly help the patient’s (young ASD kids) well-being. In this regard, the research has been initiated by developing an algorithm based on a neural network that can efficiently differentiate the brain activity of a normal young subject and an autistic young subject. In this research, Electroencephalography (EEG) data were collected from normal kids and kids with ASD from age 4 to 6. Discrete Wavelet Transform (DWT) is used for feature extraction of EEG data for valence state analysis on younger kids. It was inferred that there is a linear increase in Power Spectral Density (PSD) irrespective of age during valence state analysis of various EEG bands such as gamma, beta, alpha, and theta. When comparing the PSD of normal subjects with subjects of ASD, the PSD of ASD subjects is comparatively higher than the PSD of normal subjects. The trained network can classify the EEG data as normal subjects and subjects with ASD with good accuracy from the datasets.