Improving Motor Imagery Brain-Computer Interface Performance Through Data Screening
0
Citation
21
Reference
10
Related Paper
Keywords:
Motor Imagery
Interface (matter)
Data pre-processing
Brain Computer Interface(BCI) 기술은 뇌에서 발생한 신호를 직접 해석하여 신체 다른 기관의 표현 없이 대상의 의도를 파악하는 기술을 말한다. 실제 BCI응용의 경우 시각, 청각, 촉각 등 다양한 방법으로 단서(cue)를 제시하고 이를 기반으로 동작 할 수 있는데, 현재 많은 연구들은 BCI알고리즘의 훈련데이터와 평가데이터 사이에서 같은 종류의 단서만을 사용하여 연구를 진행하였다. 본 연구에서는 비침습형 BCI의 대표적인 방식인 EEG 기반 BCI 응용을 시각단서와 청각단서를 이용하여 평가해보았다. 본 연구의 목적은 Neurofeedback이 있는 경우와 없는 경우에 대해서 시각단서와 청각단서에 의한 Motor Imagery를 교차 성능 평가하는 것에 있다. 평가의 대상이 되는 BCI 알고리즘은 Common Spatial Pattern(CSP)과 Least Square Linear Classifier, Linear Discriminant Analysis(LDA), Support Vector Machine(SVM)을 기반으로 왼쪽 또는 오른쪽 팔을 움직이는 운동심상을 분류하며, 임상평가를 통해 실험을 진행하였다.
Motor Imagery
Sensorimotor rhythm
Cite
Citations (0)
Abstract Objective . Brain-computer interface (BCI) systems read and interpret brain activity directly from the brain. They can provide a means of communication or locomotion for patients suffering from neurodegenerative diseases or stroke. However, non-stationarity of brain activity limits the reliable transfer of the algorithms that were trained during a calibration session to real-time BCI control. One source of non-stationarity is the user’s brain response to the BCI output (feedback), for instance, whether the BCI feedback is perceived as an error by the user or not. By taking such sources of non-stationarity into account, the reliability of the BCI can be improved. Approach . In this work, we demonstrate a real-time implementation of a hybrid motor imagery BCI combining the information from the motor imagery signal and the error-related brain activity simultaneously so as to gain benefit from both sources. Main results . We show significantly improved performance in real-time BCI control across 12 participants, compared to a conventional motor imagery BCI. The significant improvement is in terms of classification accuracy, target hit rate, subjective perception of control and information-transfer rate. Moreover, our offline analyses of the recorded EEG data show that the error-related brain activity provides a more reliable source of information than the motor imagery signal. Significance . This work shows, for the first time, that the error-related brain activity classifier compared to the motor imagery classifier is more consistent when trained on calibration data and tested during online control. This likely explains why the proposed hybrid BCI allows for a more reliable means of communication or rehabilitation for patients in need.
Motor Imagery
Interface (matter)
Cite
Citations (18)
This study establishes the effectiveness of event related synchronisation (ERS) features for a system paced brain computer interface (BCI). In particular, the relationship between the duration of motor imagery (MI) and the quality of the features extracted from the ERS is investigated. To this end, two groups of users performed brief (2s) or sustained (4s) MI, and offline single trial BCIs were validated on each group based on features extracted from the EEG before, during and after MI. The BCIs were designed to recognise two intentional control tasks and a no-control state. Cross-validated results indicate that brief MI leads to more informative ERS features than sustained MI.
Motor Imagery
Interface (matter)
Cite
Citations (13)
We studied a motor-imagery brain-computer interface (MI-BCI). An MI-BCI is an interface that allows a computer to be operated by changes in brain activity that occurs when the operator imagines moving a body part. For example, with MI-BCI it is possible to assign left-hand motor-imagery to power an ON/OFF command. One of the problems with MI-BCI is its low performance, especially since MI-BCI has few commands. We aimed to improve the performance of MI-BCI by adding to the number of commands. Currently, MI-BCI has four commands based on “left hand,” “right hand,” “legs,” and “tongue” motor imagery. Therefore, we attempted to add to the number of MI-BCI commands by classifying eight kinds of brain motor-imagery activity: “no imagery,” “left hand,” “right hand,” “legs,” “both hands,” “left hand + legs,” “right hand + legs,” and “both hands + legs.” Motor imagery that involves multiple body parts, for example, “both hands,” is referred to as a multi-mental task. Multi-mental tasks involve a combination of simultaneous motor imagery, for example including the left and right hands and the legs. This makes it possible to increase the number of commands to 2N (where N is the number of body parts). Eighteen healthy males in their twenties participated in this study. The use of multi-mental tasks enabled us to improve MI-BCI performance in two out of three subjects. Multi-mental tasks can be used to add choice to MI tasks. Performance improvements using an MI-BCI were made possible by choosing MI tasks associated with high accuracy.
Motor Imagery
Interface (matter)
Cite
Citations (2)
SUMMARY The brain–computer interface (BCI) is a system to obtain information from brain signals to control computers. P300 and motor imagery tasks in electroencephalograms are the most used features for BCI. However, BCI with P300 classifies only two states and the features of the motor imagery task are too obscure to be classified easily. Therefore, we propose a method of increasing the number of classified states with high accuracy by mixed signal processing for P300 and motor imagery tasks. BCI using P300 and a motor imagery task will have a higher bit rate than conventional BCI. We design an experiment that gives four data classes, namely, control, P300, and P300 for motor imagery of the right hand or left hand. First, we confirm that P300 appears during motor imagery tasks. In addition, we investigate the best method of feature extraction. Finally, we classify four classes by means of multiclass support vector machines, and show the effectiveness of mixed signals that contain P300 and motor imagery.
Motor Imagery
Interface (matter)
Cite
Citations (4)
Motor Imagery
Interface (matter)
Cite
Citations (22)
Motor Imagery
Interface (matter)
Cite
Citations (29)
Application of Brain Computer Interface (BCI) is revolutionizing control of prosthetic or exoskeleton devices directly through human thought. A BCI is expected to classify day-to-day life activities like grabbing and lifting a glass of water. Currently, motor imagery based BCI for two closely separated muscle groups like grabbing and lifting an object has not been studied. Challenge of classifying motor imagery of these activities accurately could be solved by using individual BCI. We proposed to achieve the same by using a neural network (machine learning) classifier on high resolution (129 channel) EEG data evaluated continuously every 80ms after spatial filtering using spherical Laplacian. This study employed a motor imagery based BCI optimized for individual subjects (n=28) using EEG data of actual movement for classifying motor imagery of grab, lift and grab+lift of right forearm. A three layered neural network with two output nodes was created for classifying the motor imagery using power of 8-14 Hz band of 500 ms EEG data. This BCI was able to classify motor imagery with 95.65% accuracy. In continuous evaluation, BCI showed a True Positive Rate of 24.89% and False Positive Rate of 12.93%. The percentage of correctly classified motor imagery in each trial was 84.99%, 72.23%, 17.07% for grab, lift and combined respectively. In conclusion, the current BCI was able to classify the motor imagery of grab, lift and grab+lift successfully based on EEG of movement data without any prior training of motor imagery based on last 500ms of data.
Motor Imagery
Interface (matter)
Cite
Citations (2)
Brain-Computer Interface (BCI) is a system to obtain information from the brain signal to control computers. P300 and motor imagery task of Electroencephalogram (EEG) are mainly used features for BCI. However, BCI with P300 classifies only two states and features of motor imagery task are too obscure to be classified easily. Therefore, we propose a method to increase the number of classified states with high accuracy by mixed signal processing for P300 and motor imaginary task. BCI using P300 and motor imaginary task is going to have more bit rate than conventional BCI. We design a experiment which gives 4 classes data as control, P300, and P300 during motor imagery of right or left hand. First, we confirm that P300 appear during motor imagery task. In addition, we examine the best method for feature extraction. Finally, we classify 4 classes by multi-class Support Vector Machines, and show the efficacy of mixed signal which contain P300 and motor imagery.
Motor Imagery
Interface (matter)
SIGNAL (programming language)
Cite
Citations (0)
Multi-class EEG-based BCIs (brain-computer interfaces) usually use a set of different mental tasks to generate different commands. This study shows that, after training with a specially designed BCI paradigm using one motor imagery, humans can learn to predict the time course of band power features of the EEG signals. With this newly-obtained prediction skill, subjects can use only one motor imagery to select one of the four targets on screen
in each trial that lasts 3.4 seconds on average, which is functionally analogous to a 4-class synchronous BCI.
Motor Imagery
Cite
Citations (1)