Attention Measurement of an Autism Spectrum Disorder User Using EEG Signals: A Case Study

Autism Spectrum Disorder (ASD) is a neurodevelopmental life condition characterized by problems with social interaction, low verbal and non-verbal communication skills, repetitive restricted behavior. People ASD usually have variable attention levels because they hypersensitivity large amounts of environmental information are problem for them. Attention process that occurs at the cognitive level allows us to orient ourselves towards relevant stimuli, ignoring those not, act accordingly. This paper presents methodology based on electroencephalographic (EEG) signals measurement in 13-year-old boy diagnosed ASD. The EEG acquired an Epoc+ Brain–Computer Interface (BCI) via Emotiv Pro platform while developing several learning activities using Matlab 2019a signal processing. For this article, we propose use electrodes F3, F4, P7, P8. Then, calculate band power spectrum density detect Theta Relative Power (TRP), Alpha (ARP), Beta (BRP), Theta–Beta Ratio (TBR), Theta–Alpha (TAR), Theta/(Alpha+Beta), which features related detection neurofeedback. We train evaluate machine (ML) models these features. In study, multi-layer perceptron neural network model (MLP-NN) has best performance, AUC 0.9299, Cohen’s Kappa coefficient 0.8597, Matthews correlation 0.8602, Hamming loss 0.0701. These findings make it possible develop better scenarios according person’s needs Moreover, makes obtain quantifiable their progress reinforce perception teacher or therapist.