Improved P300-Based BCI using EEG Scalogram Images and Deep Neural Networks

Advances in Brain-Computer Interfaces (BCIs) enable persons to control physical devices and applications using their brain signals. One common approach to designing BCI systems is to use the P300 response, a positive deflection in the EEG waveform that occurs after a relevant stimulus is presented. In this paper, we propose an improved P300-based BCI using EEG scalogram images and Deep Neural Networks (DNNs). EEG scalograms is a spectrum of the EEG signals, which are used to extract relevant features for detecting P300 responses. DNNs are capable of learning complex patterns in data, making them well-suited for analyzing EEG scalogram images. We present a five-step process for building and evaluating a P300-based BCI using EEG scalogram images and DNNs. Our proposed approach promises improved accuracy and robustness in detecting P300 responses, which can enable individuals with disabilities to control devices and applications with greater ease and efficiency.