EEG Analysis for Intellectual Development Disorder

Deep learning has been shown to be able to support the analysis of imaging and signal based techniques such as electroencephalograms (EEG). In this work we apply deep learning methods in order to classify individuals with intellectual developmental disorder (IDD) using EEG signals. The dataset used contains 7 IDD subjects and 7 control subjects. EEG recordings were taken using a dry electrode system while subjects were in a resting state and while listening to music. Scalograms of the recordings were then generated using a Morlet wavelet. These scalograms were fed into a deep learning model for classification. This project compared the performance of using several fully connected and convolutional neural network (CNN) models. The highest accuracy achieved was 98.9% and 99.8% while subjects were at rest and listening to music, respectively.