Upper Limb Complex Movement Classification Using Pre-Movement EEG Signals

Abstract

A promising technology for facilitating communication and control for people with dis abilities is the brain-computer interface (BCI). Electroencephalogram (EEG) signals are frequently used in BCI systems; however, accurate classification of these signals remains challenging. This thesis presents a novel method for EEG signal classification based on spectral features. Short-time Fourier transforms (STFT) and spectral feature ex traction is used to provide an accurate classification approach for upper limb dynamic movements using pre-movement EEG signals. The proposed method was tested on a larger dataset of healthy subjects, and the performance was evaluated using different classification algorithms, including Convolutional Neural Networks (CNN) and Residual Networks (Resnet). The results show that the suggested method consistently produces high accuracy rates for all subjects and movements, with an overall accuracy of 88.7% and the highest accuracy of 100% achieved on subject 5 during movement 3 using Resnet on a privately available dataset that was compiled from 12 healthy subjects and con sisted of 5 types of upper-limb complex pre-movements that were done in 50 trails. My study extends the previous work by using a different feature extraction method and classification algorithms on a larger dataset of healthy subjects, outperforming previ ous methods. Utilizing spectral features, my method could improve the accuracy of BCI systems in various applications, including medical diagnosis, control of assistive devices, and gaming software. Furthermore, this approach could also be extended to other types of signals beyond EEG, enabling accurate classification in a broader range of applications.