Affective brain-computer interfaces (aBCIs) has important application value in the field of human-computer interaction. Electroencephalogram (EEG) has been widely concerned in the field of emotion recognition due to its advantages in time resolution, reliability and accuracy. However, the non-stationary characteristics and individual differences of EEG limit the generalization of emotion recognition model in different time and different subjects. In this paper, in order to realize the recognition of emotional states across different subjects and sessions, we proposed a new domain adaptation method, the maximum classifier difference for domain adversarial neural networks (MCD_DA). By establishing a neural network emotion recognition model, the shallow feature extractor was used to resist the domain classifier and the emotion classifier, respectively, so that the feature extractor could produce domain invariant expression, and train the decision boundary of classifier learning task specificity while realizing approximate joint distribution adaptation. The experimental results showed that the average classification accuracy of this method was 88.33% compared with 58.23% of the traditional general classifier. It improves the generalization ability of emotion brain-computer interface in practical application, and provides a new method for aBCIs to be used in practice.
Individual differences of P300 potentials lead to that a large amount of training data must be collected to construct pattern recognition models in P300-based brain-computer interface system, which may cause subjects’ fatigue and degrade the system performance. TrAdaBoost is a method that transfers the knowledge from source area to target area, which improves learning effect in the target area. Our research purposed a TrAdaBoost-based linear discriminant analysis and a TrAdaBoost-based support vector machine to recognize the P300 potentials across multiple subjects. This method first trains two kinds of classifiers separately by using the data deriving from a small amount of data from same subject and a large amount of data from different subjects. Then it combines all the classifiers with different weights. Compared with traditional training methods that use only a small amount of data from same subject or mixed different subjects’ data to directly train, our algorithm improved the accuracies by 19.56% and 22.25% respectively, and improved the information transfer rate of 14.69 bits/min and 15.76 bits/min respectively. The results indicate that the TrAdaBoost-based method has the potential to enhance the generalization ability of brain-computer interface on the individual differences.
The brain computer interface (BCI) can be used to control external devices directly through electroencephalogram (EEG) information. A multi-linear principal component analysis (MPCA) framework was used for the limitations of tensor form of multichannel EEG signals processing based on traditional principal component analysis (PCA) and two-dimensional principal component analysis (2DPCA). Based on MPCA, we used the projection of tensor-matrix to achieve the goal of dimensionality reduction and features exaction. Then we used the Fisher linear classifier to classify the features. Furthermore, we used this novel method on the BCI competitionⅡdataset 4 and BCI competitionⅣdataset 3 in the experiment. The second-order tensor representation of time-space EEG data and the third-order tensor representation of time-space-frequency EEG data were used. The best results that were superior to those from other dimensionality reduction methods were obtained by much debugging on parameter P and testQ. For two-order tensor, the highest accuracy rates could be achieved as 81.0% and 40.1%, and for three-order tensor, the highest accuracy rates were 76.0% and 43.5%, respectively.
With the breakthroughs of digitization, artificial intelligence and other technologies and the gradual expansion of application fields, more and more studies have been conducted on the application of digital intelligence technologies such as exoskeleton robots, brain-computer interface, and spinal cord neuromodulation to improve or compensate physical function after spinal cord injury (SCI) and improve self-care ability and quality of life of patients with SCI. The development of digital intelligent rehabilitation technology provides a new application platform for the functional reconstruction after SCI, and the digital and intelligentized rehabilitation technology has broad application prospects in the clinical rehabilitation treatment after SCI. This article elaborates on the current status of exoskeleton robots, brain-computer interface technology, and spinal cord neuromodulation technology for functional recovery after SCI.
Brain-computer interface (BCI) based on functional near-infrared spectroscopy (fNIRS) is a new-type human-computer interaction technique. To explore the separability of fNIRS signals in different motor imageries on the single limb, the study measured the fNIRS signals of 15 subjects (amateur football fans) during three different motor imageries of the right foot (passing, stopping and shooting). And the correlation coefficient of the HbO signal during different motor imageries was extracted as features for the input of a three-classification model based on support vector machines. The results found that the classification accuracy of the three motor imageries of the right foot was 78.89%±6.161%. The classification accuracy of the two-classification of motor imageries of the right foot, that is, passing and stopping, passing and shooting, and stopping and shooting was 85.17%±4.768%, 82.33%±6.011%, and 89.33%±6.713%, respectively. The results demonstrate that the fNIRS of different motor imageries of the single limb is separable, which is expected to add new control commands to fNIRS-BCI and also provide a new option for rehabilitation training and control peripherals for unilateral stroke patients. Besides, the study also confirms that the correlation coefficient can be used as an effective feature to classify different motor imageries.
Electroencephalogram (EEG) classification for brain-computer interface (BCI) is a new way of realizing human-computer interreaction. In this paper the application of semi-supervised sparse representation classifier algorithms based on help training to EEG classification for BCI is reported. Firstly, the correlation information of the unlabeled data is obtained by sparse representation classifier and some data with high correlation selected. Secondly, the boundary information of the selected data is produced by discriminative classifier, which is the Fisher linear classifier. The final unlabeled data with high confidence are selected by a criterion containing the information of distance and direction. We applied this novel method to the three benchmark datasets, which were BCIⅠ, BCIⅡ_Ⅳ and USPS. The classification rate were 97%,82% and 84.7%, respectively. Moreover the fastest arithmetic rate was just about 0.2 s. The classification rate and efficiency results of the novel method are both better than those of S3VM and SVM, proving that the proposed method is effective.
In the research of non-invasive brain-computer interface (BCI), independent component analysis (ICA) has been considered as a promising method of electroencephalogram (EEG) preprocessing and feature enhancement. However, there have been few investigations and implements about online ICA-BCI system up till now. This paper reports the investigation of the ICA-based motor imagery BCI (MIBCI) system, combining the characteristics of unsupervised learning of ICA and event-related desynchronization (ERD) related to motor imagery. We constructed a simple and practical method of ICA spatial filter calculation and discriminate criterion of three-type motor imageries in the study. To validate the online performance of proposed algorithms, an ICA-MIBCI experimental system was fully established based on NeuroScan EEG amplifier and VC++ platform. Four subjects participated in the experiment of MIBCI testing and two of them took part in the online experiment. The average classification accuracies of the three-type motor imageries reached 89.78% and 89.89% in the offline and online testing, respectively. The experimental results showed that the proposed algorithm produced high classification accuracy and required less time consumption, which would have a prospect of cross platform application.
Brain-computer interface (BCI) can be summarized as a system that uses online brain information to realize communication between brain and computer. BCI has experienced nearly half a century of development, although it now has a high degree of awareness in the public, but the application of BCI in the actual scene is still very limited. This collection invited some BCI teams in China to report their efforts to promote BCI from laboratory to real scene. This paper summarizes the main contents of the invited papers, and looks forward to the future of BCI.
High-density channels are often used to acquire electroencephalogram (EEG) spatial information in different cortical regions of the brain in brain-computer interface (BCI) systems. However, applying excessive channels is inconvenient for signal acquisition, and it may bring artifacts. To avoid these defects, the common spatial pattern (CSP) algorithm was used for channel selection and a selection criteria based on norm-2 is proposed in this paper. The channels with the highest M scores were selected for the purpose of using fewer channels to acquire similar rate with high density channels. The DatasetⅢa from BCI competition 2005 were used for comparing the classification accuracies of three motor imagery between whole channels and the selected channels with the present proposed method. The experimental results showed that the classification accuracies of three subjects using the 20 channels selected with the present method were all higher than the classification accuracies using all 60 channels, which convinced that our method could be more effective and useful.
Brain-computer interaction (BCI) is a transformative human-computer interaction, which aims to bypass the peripheral nerve and muscle system and directly convert the perception, imagery or thinking activities of cranial nerves into actions for further improving the quality of human life. Magnetoencephalogram (MEG) measures the magnetic field generated by the electrical activity of neurons. It has the unique advantages of non-contact measurement, high temporal and spatial resolution, and convenient preparation. It is a new BCI driving signal. MEG-BCI research has important brain science significance and potential application value. So far, few documents have elaborated the key technical issues involved in MEG-BCI. Therefore, this paper focuses on the key technologies of MEG-BCI, and details the signal acquisition technology involved in the practical MEG-BCI system, the design of the MEG-BCI experimental paradigm, the MEG signal analysis and decoding key technology, MEG-BCI neurofeedback technology and its intelligent method. Finally, this paper also discusses the existing problems and future development trends of MEG-BCI. It is hoped that this paper will provide more useful ideas for MEG-BCI innovation research.