Logistic Regression based Feature Selection and Two-Stage Detection for EEG based Motor Imagery Classification

Electroencephalogram (EEG) based motor imagery (MI) classification requires efficient feature extraction and consistent accuracy for reliable brain-computer interface (BCI) systems. Achieving consistent accuracy in EEGMI classification is still big challenge according to the nature of EEG signal which is subject dependent. To address this problem, we propose a feature selection scheme based on Logistic Regression (LRFS) and two-stage detection (TSD) in channel instantiation approach. In TSD scheme, Linear Discriminant Analysis was utilized in first-stage detection; while Gradient Boosted Tree and k-Nearest Neighbor in second-stage detection. To evaluate the proposed method, two publicly available datasets, BCI competition III-Dataset IVa and BCI competition IV-Dataset 2a, were used. Experimental results show that the proposed method yielded excellent accuracy for both datasets with 95.21% and 94.83%, respectively. These results indicated that the proposed method has consistent accuracy and is promising for reliable BCI systems.

[1]  Deron Liang,et al.  The effect of feature selection on financial distress prediction , 2015, Knowl. Based Syst..

[2]  Jianjun Meng,et al.  Simultaneously Optimizing Spatial Spectral Features Based on Mutual Information for EEG Classification , 2015, IEEE Transactions on Biomedical Engineering.

[3]  Toshihisa Tanaka,et al.  Robust Averaging of Covariances for EEG Recordings Classification in Motor Imagery Brain-Computer Interfaces , 2017, Neural Computation.

[4]  Marcos Dipinto,et al.  Discriminant analysis , 2020, Predictive Analytics.

[5]  Mohammed Imamul Hassan Bhuiyan,et al.  Classification of motor imagery movements using multivariate empirical mode decomposition and short time Fourier transform based hybrid method , 2016 .

[6]  Wonzoo Chung,et al.  Frequency-Optimized Local Region Common Spatial Pattern Approach for Motor Imagery Classification , 2019, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[7]  Tao Zhang,et al.  Structural and functional correlates of motor imagery BCI performance: Insights from the patterns of fronto-parietal attention network , 2016, NeuroImage.

[8]  Xingyu Wang,et al.  Temporally Constrained Sparse Group Spatial Patterns for Motor Imagery BCI , 2019, IEEE Transactions on Cybernetics.

[9]  Hongxin Zhang,et al.  A self-adaptive frequency selection common spatial pattern and least squares twin support vector machine for motor imagery electroencephalography recognition , 2018, Biomed. Signal Process. Control..

[10]  Mark H. Johnson,et al.  An EEG study on the somatotopic organisation of sensorimotor cortex activation during action execution and observation in infancy , 2015, Developmental Cognitive Neuroscience.

[11]  Bin He,et al.  EEG Source Imaging Enhances the Decoding of Complex Right-Hand Motor Imagery Tasks , 2016, IEEE Transactions on Biomedical Engineering.

[12]  Girijesh Prasad,et al.  Covariate shift estimation based adaptive ensemble learning for handling non-stationarity in motor imagery related EEG-based brain-computer interface , 2018, Neurocomputing.

[13]  R. Leeb,et al.  BCI Competition 2008 { Graz data set B , 2008 .

[14]  Girish Kumar Singh,et al.  Sub-band classification of decomposed single event-related potential co-variants for multi-class brain–computer interface: a qualitative and quantitative approach , 2016 .

[15]  Rohit Bose,et al.  Detection of epileptic seizure and seizure-free EEG signals employing generalised S -transform , 2017 .

[16]  Isabelle Bloch,et al.  Subject-specific time-frequency selection for multi-class motor imagery-based BCIs using few Laplacian EEG channels , 2017, Biomed. Signal Process. Control..

[17]  Na Zhang,et al.  Hidden-layer visible deep stacking network optimized by PSO for motor imagery EEG recognition , 2017, Neurocomputing.

[18]  Yanhui Xu,et al.  Classification Based on Multilayer Extreme Learning Machine for Motor Imagery Task from EEG Signals , 2016, BICA.

[19]  Xingyu Wang,et al.  An Optimized Channel Selection Method Based on Multifrequency CSP-Rank for Motor Imagery-Based BCI System , 2019, Comput. Intell. Neurosci..

[20]  Yang Yu,et al.  Toward brain-actuated car applications: Self-paced control with a motor imagery-based brain-computer interface , 2016, Comput. Biol. Medicine.

[21]  Eric Granger,et al.  Multiple instance learning: A survey of problem characteristics and applications , 2016, Pattern Recognit..

[22]  Shuning Yang,et al.  Motor Imagery EEG Classification Based on Decision Tree Framework and Riemannian Geometry , 2019, Comput. Intell. Neurosci..

[23]  Anina N. Rich,et al.  Multimodal functional imaging of motor imagery using a novel paradigm , 2013, NeuroImage.

[24]  Shaun Boe,et al.  Specific Brain Lesions Impair Explicit Motor Imagery Ability: A Systematic Review of the Evidence. , 2016, Archives of physical medicine and rehabilitation.

[25]  Mohammad Reza Karami Mollaei,et al.  A novel method for classification of BCI multi-class motor imagery task based on Dempster-Shafer theory , 2019, Inf. Sci..

[26]  Roberto Antonio Vázquez,et al.  Evaluating the effect of the cutoff frequencies during the pre-processing stage of motor imagery EEG signals classification , 2019, Biomed. Signal Process. Control..

[27]  Yu Zhang,et al.  Sparse Group Representation Model for Motor Imagery EEG Classification , 2019, IEEE Journal of Biomedical and Health Informatics.

[28]  Mario Ignacio Chacon Murguia,et al.  Classification of multiple motor imagery using deep convolutional neural networks and spatial filters , 2019, Appl. Soft Comput..

[29]  Ying Wen,et al.  Motor imagery EEG signals analysis based on Bayesian network with Gaussian distribution , 2014, Neurocomputing.

[30]  Tao Yang,et al.  Automated classification of neonatal amplitude-integrated EEG based on gradient boosting method , 2016, Biomed. Signal Process. Control..

[31]  Onder Aydemir,et al.  A polynomial fitting and k-NN based approach for improving classification of motor imagery BCI data , 2010, Pattern Recognit. Lett..

[32]  Pedro J. García-Laencina,et al.  Exploring dimensionality reduction of EEG features in motor imagery task classification , 2014, Expert Syst. Appl..

[33]  Sofien Gannouni,et al.  A dynamic and self-adaptive classification algorithm for motor imagery EEG signals , 2019, Journal of Neuroscience Methods.

[34]  Hua Wang,et al.  Detection of motor imagery EEG signals employing Naïve Bayes based learning process , 2016 .

[35]  Prasant Kumar Pattnaik,et al.  Brain Computer Interface issues on hand movement , 2018, J. King Saud Univ. Comput. Inf. Sci..

[36]  Gabriela Castellano,et al.  Space-time recurrences for functional connectivity evaluation and feature extraction in motor imagery brain-computer interfaces , 2019, Medical & Biological Engineering & Computing.

[37]  G. Pfurtscheller,et al.  The BCI competition III: validating alternative approaches to actual BCI problems , 2006, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[38]  Zhang Hong-xin,et al.  Recognition of motor imagery tasks for BCI using CSP and chaotic PSO twin SVM , 2017 .

[39]  Hamid Mirvaziri,et al.  Improvement of EEG-based motor imagery classification using ring topology-based particle swarm optimization , 2017, Biomed. Signal Process. Control..

[40]  Jasmin Kevric,et al.  Biomedical Signal Processing and Control , 2016 .

[41]  Jyoti Singh Kirar,et al.  A combination of spectral graph theory and quantum genetic algorithm to find relevant set of electrodes for motor imagery classification , 2020, Appl. Soft Comput..

[42]  Rongrong Fu,et al.  Improvement Motor Imagery EEG Classification Based on Regularized Linear Discriminant Analysis , 2019, Journal of Medical Systems.

[43]  David Lee,et al.  Filter Bank Regularized Common Spatial Pattern Ensemble for Small Sample Motor Imagery Classification , 2018, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[44]  Girijesh Prasad,et al.  An Automatic Subject Specific Intrinsic Mode Function Selection for Enhancing Two-Class EEG-Based Motor Imagery-Brain Computer Interface , 2019, IEEE Sensors Journal.

[45]  Xiaomu Song,et al.  Improving brain-computer interface classification using adaptive common spatial patterns , 2015, Comput. Biol. Medicine.

[46]  Lei Sun,et al.  A contralateral channel guided model for EEG based motor imagery classification , 2018, Biomed. Signal Process. Control..

[47]  Po-Lei Lee,et al.  Recognition of Motor Imagery Electroencephalography Using Independent Component Analysis and Machine Classifiers , 2005, Annals of Biomedical Engineering.

[48]  Guanghua Xu,et al.  Classification of single-trial motor imagery EEG by complexity regularization , 2017, Neural Computing and Applications.

[49]  Li Zhang,et al.  Differential evolution algorithm as a tool for optimal feature subset selection in motor imagery EEG , 2017, Expert Syst. Appl..