Year: 2025
Type: article
Abstract: Brain-Computer Interfaces (BCIs) have opened new doors for individuals with motor impairments by enabling direct communication with computers without requiring physical movement. One of the most promising applications of BCI technology is the P300 speller, which allows users to select letters by detecting brain signals known as P300 event-related potentials. This study introduces a method to enhance the performance of the P300 speller using Convolutional Neural Networks (CNNs). The research is based on data from the BCI Competition III Dataset II, and we have used modern EEG signal processing techniques to train our model. The results show that the CNN-based model provides better accuracy compared to traditional machine learning approaches, making virtual keyboards more accessible and practical for those with disabilities. We also explore some of the challenges faced by current BCI spellers and suggest potential improvements that can be implemented in future research. The ultimate goal of this study is to contribute to the development of a reliable and efficient communication system for individuals with severe disabilities.
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W4410128453