Towards accurate BCI-based classification via deep learning: a study of hand movement application

Bowen Tan

doi:10.1117/12.2674809

23 May 2023 Towards accurate BCI-based classification via deep learning: a study of hand movement application

Bowen Tan

Proceedings Volume 12604, International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2022); 126044U (2023) https://doi.org/10.1117/12.2674809
Event: 2nd International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2022), 2022, Guangzhou, China

Abstract

The application of brain-computer interface (BCI) has been booming recently, the advanced artificial intelligence techniques can be applied to improve the performance of BCI applications. In this paper, an electroencephalogram-based hand movement application has been investigated. This is a three-classes classification task obtaining right, left direction movement, and hold status. We proposed two deep learning (DL)-based frameworks, one is convolution neural networks-based and the other one is long short-term memory (LSTM)-based architecture. The results show that DL-based models achieved an accuracy of 62% and 65% respectively, which is better than traditional machine learning (ML) algorithms. Furthermore, we also split the dataset into different percentages to explore the impact of the number of training data to the performance. Eventually, results indicate that the larger the data set the better the training results, and this applies to both DL-based models and ML-based models. Additionally, we used the above two DL-based frameworks to deal with participants classification.

Citation Download Citation

Bowen Tan "Towards accurate BCI-based classification via deep learning: a study of hand movement application", Proc. SPIE 12604, International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2022), 126044U (23 May 2023); https://doi.org/10.1117/12.2674809

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available