scholarly journals Simultaneously Spatiospectral Pattern Learning and Contaminated Trial Pruning for Electroencephalography-Based Brain Computer Interface

Symmetry ◽  
2020 ◽  
Vol 12 (9) ◽  
pp. 1387
Author(s):  
Chun-Ping Shieh ◽  
Shih-Hung Yang ◽  
Yu-Shun Liu ◽  
Yun-Ting Kuo ◽  
Yu-Chun Lo ◽  
...  
Keyword(s):  
Mutual Information ◽  
Motor Imagery ◽  
Spatial Patterns ◽  
Conditional Probability Density ◽  
Approximation Technique ◽  
Contamination Level ◽  
Spatial Filters ◽  
Frequency Bands ◽  
Computer Interfaces ◽  
The Difference

Electroencephalography (EEG)-based brain computer interfaces (BCIs) translate motor imagery commands into the movements of an external device (e.g., a robotic arm). The automatic design of spectral and spatial filters is a challenging task, as the frequency bands of the spectral filters must be predefined by previously published studies and given that they may be affected during trials by artifacts and improper motor imagery (MI). This study aimed to eliminate the contaminated trials automatically during classifier training, and to simultaneously learn the spectral and spatial patterns without the need for predefined frequency bands. Compared with previous studies that measured the discriminative power of a frequency band based on mutual information, this study determined the difference of the class conditional probability density function between two MI classes. This information was further shared to measure the contamination level of the trial that simplified the computation. A particle-based approximation technique iteratively constructed a filter bank that extracted discriminative features, and simultaneously removed potentially contaminated trials. The particle weight was estimated by an analysis of variance F-test instead of mutual information as commonly used in previous studies. The experimental results of a publicly available dataset revealed that the proposed method outperformed the other BCI in terms of the classification accuracy. Asymmetrical spatial patterns were found on left- versus right-hand MI classifications. The learnt spectral and spatial patterns were consistent with prior neurophysiological knowledge.

Selecting Filter Range of Hybrid Brain-Computer Interfaces by Mutual Information

2014 ◽  
Vol 981 ◽  
pp. 171-174 ◽  
Author(s):  
Li Wang ◽  
Xiong Zhang ◽  
Xue Fei Zhong ◽  
Zhao Wen Fan
Keyword(s):  
Mutual Information ◽  
Motor Imagery ◽  
Support Vector ◽  
Frequency Range ◽  
Fixed Frequency ◽  
Eeg Signals ◽  
Optimum Frequency ◽  
Computer Interfaces ◽  
Narrow Frequency Range ◽  
Better Than

The hybrid brain-computer interface (BCI) based on electroencephalography (EEG) become more and more popular. Motor imagery, steady state visual evoked potentials (SSVEPs) and P300 are main training Paradigms. In our previous research, BCI systems based on motor imagery can be extended by speech imagery. However, noise and artifact may be produced by different mental tasks and EEG signals are also different among users, so the classification accuracy can be improved by selecting optimum frequency range for each user. Mutual information (MI) is usually used to choose optimal features. After extracted the features from each narrow frequency range of EEG by common spatial patterns (CSP), the features are assessed by MI. Then, the optimum frequency range can be acquired. The final classification results are calculated by support vector machine (SVM). The average result of optimum frequency range from seven subjects is better than the result of a fixed frequency range.

scholarly journals Across-subject offline decoding of motor imagery from MEG and EEG

2018 ◽  
Author(s):  
Hanna-Leena Halme ◽  
Lauri Parkkonen
Keyword(s):  
Logistic Regression ◽  
Motor Imagery ◽  
Spatial Patterns ◽  
Computer Interfaces ◽  
The Common ◽  
Calibration Time ◽  
Decoding Accuracy ◽  
And Training ◽  
Healthy Participants ◽  
Passive Movements

AbstractLong calibration time hinders the feasibility of brain-computer interfaces (BCI). If other subjects’ data were used for training the classifier, BCI-based neurofeedback practice could start without the initial calibration. Here, we compare methods for inter-subject decoding of left- vs. right-hand motor imagery (MI) from MEG and EEG.Six methods were tested on data involving MEG and EEG measurements of healthy participants. Only subjects with good within-subject accuracies were selected for inter-subject decoding. Three methods were based on the Common Spatial Patterns (CSP) algorithm, and three others on logistic regression with l1 - or l2,1 -norm regularization. The decoding accuracy was evaluated using 1) MI and 2) passive movements (PM) for training, separately for MEG and EEG.When the classifier was trained by MI, the best accuracies across subjects (mean 70.6% for MEG, 67.7% for EEG) were obtained using multi-task learning (MTL) with logistic regression and l2,1-norm regularization. MEG yielded slightly better average accuracies than EEG. When PM were used for training, none of the inter-subject methods yielded above chance level (58.7%) accuracy.In conclusion, MTL and training with other subject’s MI is efficient for inter-subject decoding of MI. Passive movements of other subjects are likely suboptimal for training the MI classifiers.

Transfer Learning Based on Regularized Common Spatial Patterns Using Cosine Similarities of Spatial Filters for Motor-Imagery BCI

2019 ◽  
Vol 28 (07) ◽  
pp. 1950123 ◽  
Author(s):  
Yilu Xu ◽  
Qingguo Wei ◽  
Hua Zhang ◽  
Ronghua Hu ◽  
Jizhong Liu ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Motor Imagery ◽  
Spatial Patterns ◽  
Covariance Matrices ◽  
Training Data ◽  
Spatial Filters ◽  
Data Set ◽  
Common Spatial Patterns ◽  
Public Data ◽  
Target Subject

In motor-imagery brain–computer interface (BCI), transfer learning based on the framework of regularized common spatial patterns (RCSP) can make full use of the training data derived from other subjects to reduce calibration time for a new subject. Covariance matrices are commonly used to estimate the difference between subjects. However, the classification performances vary greatly depending on different assumptions of the distribution of covariance matrices. Therefore, to directly observe the variations of the target subject’s features after transferring, we neglect the distribution of covariance matrices and instead compare cosine similarities of spatial filters between the target subject and the composite subject whose data comes from the target subject and a source subject. Two RCSP algorithms based on cosine measure are proposed to use the samples of all source subjects and most useful source subjects, respectively. Experiments on one public data set from BCI competition show that our proposed approaches significantly improve the classification performances compared to the conventional CSP algorithm in almost every case, based on a small training set.

scholarly journals Learning Common Time-Frequency-Spatial Patterns for Motor Imagery Classification

2021 ◽  
pp. 1-1
Author(s):  
Yangyang Miao ◽  
Jing Jin ◽  
Ian Daly ◽  
Cili Zuo ◽  
Xingyu Wang ◽  
...  
Keyword(s):  
Motor Imagery ◽  
Spatial Patterns ◽  
Time Frequency ◽  
Common Time

scholarly journals Source Detection and Functional Connectivity of the Sensorimotor Cortex during Actual and Imaginary Limb Movement: A Preliminary Study on the Implementation of eConnectome in Motor Imagery Protocols

2012 ◽  
Vol 2012 ◽  
pp. 1-10 ◽  
Author(s):  
Alkinoos Athanasiou ◽  
Chrysa Lithari ◽  
Konstantina Kalogianni ◽  
Manousos A. Klados ◽  
Panagiotis D. Bamidis
Keyword(s):  
Functional Connectivity ◽  
Motor Imagery ◽  
Sensorimotor Cortex ◽  
Mu Rhythm ◽  
Source Detection ◽  
Motor Execution ◽  
Motor Tasks ◽  
Computer Interfaces ◽  
Preliminary Study ◽  
Real Hand

Introduction. Sensorimotor cortex is activated similarly during motor execution and motor imagery. The study of functional connectivity networks (FCNs) aims at successfully modeling the dynamics of information flow between cortical areas.Materials and Methods. Seven healthy subjects performed 4 motor tasks (real foot, imaginary foot, real hand, and imaginary hand movements), while electroencephalography was recorded over the sensorimotor cortex. Event-Related Desynchronization/Synchronization (ERD/ERS) of the mu-rhythm was used to evaluate MI performance. Source detection and FCNs were studied with eConnectome.Results and Discussion. Four subjects produced similar ERD/ERS patterns between motor execution and imagery during both hand and foot tasks, 2 subjects only during hand tasks, and 1 subject only during foot tasks. All subjects showed the expected brain activation in well-performed MI tasks, facilitating cortical source estimation. Preliminary functional connectivity analysis shows formation of networks on the sensorimotor cortex during motor imagery and execution.Conclusions. Cortex activation maps depict sensorimotor cortex activation, while similar functional connectivity networks are formed in the sensorimotor cortex both during actual and imaginary movements. eConnectome is demonstrated as an effective tool for the study of cortex activation and FCN. The implementation of FCN in motor imagery could induce promising advancements in Brain Computer Interfaces.

scholarly journals User Experience May be Producing Greater Heart Rate Variability than Motor Imagery Related Control Tasks during the User-System Adaptation in Brain-Computer Interfaces

2016 ◽  
Vol 7 ◽  
Author(s):  
Luz M. Alonso-Valerdi ◽  
David A. Gutiérrez-Begovich ◽  
Janet Argüello-García ◽  
Francisco Sepulveda ◽  
Ricardo A. Ramírez-Mendoza
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Motor Imagery ◽  
User Experience ◽  
Brain Computer Interfaces ◽  
Computer Interfaces ◽  
System Adaptation

Optimization of Model Training Based on Iterative Minimum Covariance Determinant In Motor-Imagery BCI

2021 ◽  
pp. 2150030
Author(s):  
Jing Jin ◽  
Hua Fang ◽  
Ian Daly ◽  
Ruocheng Xiao ◽  
Yangyang Miao ◽  
...  
Keyword(s):  
Motor Imagery ◽  
Classification Performance ◽  
Minimum Covariance Determinant ◽  
Fisher Score ◽  
Practical Applications ◽  
Computer Interfaces ◽  
Highly Sensitive ◽  
Feature Optimization ◽  
The Common ◽  
Model Training

The common spatial patterns (CSP) algorithm is one of the most frequently used and effective spatial filtering methods for extracting relevant features for use in motor imagery brain–computer interfaces (MI-BCIs). However, the inherent defect of the traditional CSP algorithm is that it is highly sensitive to potential outliers, which adversely affects its performance in practical applications. In this work, we propose a novel feature optimization and outlier detection method for the CSP algorithm. Specifically, we use the minimum covariance determinant (MCD) to detect and remove outliers in the dataset, then we use the Fisher score to evaluate and select features. In addition, in order to prevent the emergence of new outliers, we propose an iterative minimum covariance determinant (IMCD) algorithm. We evaluate our proposed algorithm in terms of iteration times, classification accuracy and feature distribution using two BCI competition datasets. The experimental results show that the average classification performance of our proposed method is 12% and 22.9% higher than that of the traditional CSP method in two datasets ([Formula: see text]), and our proposed method obtains better performance in comparison with other competing methods. The results show that our method improves the performance of MI-BCI systems.

Sign in / Sign up

Export Citation Format

Share Document