Learning Invariant Patterns Based on a Convolutional Neural Network and Big Electroencephalography Data for Subject-Independent P300 Brain-Computer Interfaces

Wei Gao,Tianyou Yu,Jin-Gang Yu,Zhenghui Gu,Kendi Li,Yong Huang,Zhu Liang Yu,Yuanqing Li

DOI: https://doi.org/10.1109/tnsre.2021.3083548

IF: 4.9

2021-01-01

IEEE Transactions on Neural Systems and Rehabilitation Engineering

Abstract:A brain-computer interface (BCI) measures and analyzes brain activity and converts this activity into computer commands to control external devices. In contrast to traditional BCIs that require a subject-specific calibration process before being operated, a subject-independent BCI learns a subject-independent model and eliminates subject-specific calibration for new users. However, building subject-independent BCIs remains difficult because electroencephalography (EEG) is highly noisy and varies by subject. In this study, we propose an invariant pattern learning method based on a convolutional neural network (CNN) and big EEG data for subject-independent P300 BCIs. The CNN was trained using EEG data from a large number of subjects, allowing it to extract subject-independent features and make predictions for new users. We collected EEG data from 200 subjects in a P300-based spelling task using two different types of amplifiers. The offline analysis showed that almost all subjects obtained significant cross-subject and cross-amplifier effects, with an average accuracy of more than 80%. Furthermore, more than half of the subjects achieved accuracies above 85%. These results indicated that our method was effective for building a subject-independent P300 BCI, with which more than 50% of users could achieve high accuracies without subject-specific calibration.

engineering, biomedical,rehabilitation

What problem does this paper attempt to address?

The problem this paper attempts to address is the development of a cross-subject P300 brain-computer interface (BCI) system that does not require specific calibration for each user. Traditional BCI systems typically require a specific calibration process for each user, which is not only time-consuming and inconvenient but also limits the widespread use of BCI in practical applications. This paper proposes an invariant pattern learning method based on convolutional neural networks (CNN) and large-scale EEG data, aiming to establish a P300 BCI system that can be applied to different users without specific calibration.

Specifically, this study addresses the following key issues:
1. **High noise and individual differences**: EEG signals are highly noisy and vary from person to person, making it very difficult to construct a cross-subject BCI system.
2. **Reducing or eliminating calibration time**: By using large-scale EEG data to train the CNN model, cross-subject features are extracted, thereby reducing or eliminating the calibration time required before new users use the BCI system.
3. **Improving system performance**: Through cross-subject and cross-amplifier data analysis, the accuracy and robustness of the BCI system are improved, enabling it to perform well on different users and different devices.

The research results show that this method achieved significant cross-subject and cross-amplifier effects in most users, with an average accuracy rate of over 80%, and more than half of the users achieved an accuracy rate of over 85%. These results indicate that by using large-scale training data, it is possible to effectively establish a cross-subject P300 BCI system without specific calibration, thereby significantly enhancing the application prospects of P300 BCI.