Optimization of Model Training Based on Iterative Minimum Covariance Determinant In Motor-Imagery BCI
Abstract
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 (), 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.
References
- 1. , Brain–computer interfaces using sensorimotor rhythms: Current state and future perspectives, IEEE Trans. Biomed. Eng. 61(5) (2014) 1425–1435. Crossref, Medline, Web of Science, Google Scholar
- 2. , The study of generic model set for reducing calibration time in P300-based brain–computer interface, IEEE Trans. Neural Syst. Rehab. Eng. 28(1) (2020) 3–12. Crossref, Medline, Web of Science, Google Scholar
- 3. , Brain–computer interface technologies: From signal to action, Rev. Neurosci. 24(5) (2013) 537–552. Crossref, Medline, Web of Science, Google Scholar
- 4. , Temporal modulation of steady-state visual evoked potentials, Int. J. Neural Syst. 29(3) (2019) 1850050. Link, Web of Science, Google Scholar
- 5. , Design of assistive wheelchair system directly steered by human thoughts, Int. J. Neural Syst. 23(3) (2013) 1350013. Link, Web of Science, Google Scholar
- 6. , Combined corticospinal and reticulospinal effects on upper limb muscles, Neurosci. Lett. 561 (2014) 30–34. Crossref, Medline, Web of Science, Google Scholar
- 7. , Wearable technology for patients with brain and spinal cord injuries, Rev. Neurosci. 28(8) (2017) 913–920. Crossref, Medline, Web of Science, Google Scholar
- 8. , Subject transfer BCI based on composite local temporal correlation common spatial pattern, Comput. Biol. Med. 64 (2015) 1–11. Crossref, Medline, Web of Science, Google Scholar
- 9. , A three-dimensional microelectrode array to generate virtual electrodes for epiretinal prosthesis based on a modeling study, Int. J. Neural Syst. 30(3) (2020) 2050006. Link, Web of Science, Google Scholar
- 10. , A training data-driven canonical correlation analysis algorithm for designing spatial filters to enhance performance of SSVEP-based BCIs, Int. J. Neural Syst. 30(5) (2020) 2050020. Link, Web of Science, Google Scholar
- 11. , Internal feature selection method of CSP based on L1-Norm and Dempster-Shafer theory, IEEE Trans. Neural Netw. Learn. Syst., Early acess,
Aug. 24, 2020 , https://doi.org/10.1109/TNNLS.2020.3015505. Google Scholar - 12. , Towards correlation-based time window selection method for motor imagery BCIs, Neural Netw. 102 (2018) 87–95. Crossref, Medline, Web of Science, Google Scholar
- 13. , Developing a novel tactile P300 brain–computer interface with a cheeks-stim paradigm, IEEE Trans. Biomed. Eng. 67(9) (2020) 2585–2593. Crossref, Medline, Web of Science, Google Scholar
- 14. , Synchronization of slow cortical rhythms during motor imagery-based brain–machine interface control, Int. J. Neural Syst. 29(5) (2019) 1850045. Link, Web of Science, Google Scholar
- 15. , A P300 brain–computer interface based on a modification of the mismatch negativity paradigm, Int. J. Neural Syst. 25(3) (2015) 1550011. Link, Web of Science, Google Scholar
- 16. , Optimizing spatial spectral patterns jointly with channel configuration for brain-computer interface, Neurocomputing 104 (2013) 115–126. Crossref, Web of Science, Google Scholar
- 17. , Gross motor ability predicts response to upper extremity rehabilitation in chronic stroke, Behav. Brain Res. 333 (2017) 314–322. Crossref, Medline, Web of Science, Google Scholar
- 18. , Upper limb movement classification via electromyographic signals and an enhanced probabilistic network, J. Med. Syst. 44(10) (2020) 176. Crossref, Medline, Web of Science, Google Scholar
- 19. , Identifying suitable brain regions and trial size segmentation for positive/negative emotion recognition, Int. J. Neural Syst. 29(2) (2019) 1850044. Link, Web of Science, Google Scholar
- 20. , The influence of visual attention on the performance of a novel tactile P300 brain–computer interface with cheeks-stim paradigm, Int. J. Neural Syst. 31 (2021) 2150004. Link, Web of Science, Google Scholar
- 21. , Pre-Stimulus sensorimotor rhythms influence brain–computer interface classification performance, IEEE Trans. Neural Syst. Rehab. Eng. 20(5) (2012) 653–662. Crossref, Medline, Web of Science, Google Scholar
- 22. , Nonspecific visuospatial imagery as a novel mental task for online EEG-based BCI control, Int. J. Neural Syst. 30(6) (2020) 2050026. Link, Web of Science, Google Scholar
- 23. , Study of the functional brain connectivity and lower-limb motorimagery performance after transcranial direct current stimulation, Int. J. Neural Syst. 30(8) (2020) 2050038. Link, Web of Science, Google Scholar
- 24. , A novel methodology for extracting and evaluating therapeutic movements in game-based motion capture rehabilitation systems, J. Med. Syst. 42(12) (2018) 255. Crossref, Medline, Web of Science, Google Scholar
- 25. , Predicting improved daily use of the more affected arm poststroke following constraint-induced movement therapy, Phys. Therapy 99(12) (2019) 1667–1678. Crossref, Medline, Web of Science, Google Scholar
- 26. , Neurolight: A deep learning neural interface for cortical visual prostheses, Int. J. Neural Syst. 30(9) (2020) 2050045. Link, Web of Science, Google Scholar
- 27. , Computer-aided prediction of extent of motor recovery following constraint-induced movement therapy in chronic stroke, Behav. Brain Res. 329 (2017) 191–199. Crossref, Medline, Web of Science, Google Scholar
- 28. , Regularized group sparse discriminant analysis for P300-based brain–computer interface, Int. J. Neural Syst. 29(6) (2019) 1950002. Link, Web of Science, Google Scholar
- 29. , Increasing N200 potentials via visual stimulus depicting humanoid robot behavior, Int. J. Neural Syst. 26(1) (2016) 1550039. Link, Web of Science, Google Scholar
- 30. , An ERP-based BCI using an oddball paradigm with different faces and reduced errors in critical functions, Int. J. Neural Syst. 24(8) (2014) 1450027. Link, Web of Science, Google Scholar
- 31. B. Blankertz, R. Tomioka, S. Lemm, M. Kawanabe and K.-R. Mueller, Optimizing spatial filters for robust EEG single-trial analysis, IEEE Signal Process. Mag. 25(1) (2008) 41–56. Google Scholar
- 32. , Optimizing the channel selection and classification accuracy in EEG-based BCI, IEEE Trans. Biomed. Eng. 58(6) (2011) 1865–1873. Crossref, Medline, Web of Science, Google Scholar
- 33. , Designing optimal spatial filters for single-trial EEG classification in a movement task, Clin. Neurophysiol. 110(5) (1999) 787–798. Crossref, Medline, Web of Science, Google Scholar
- 34. , Potential pitfalls of widely used implementations of common spatial patterns, Annual International Conference of the IEEE Engineering in Medicine and Biology Society.
20–24 July, 2020 ,Montreal, Canada , pp. 196–199. Google Scholar - 35. , Wavelet methodology to improve single unit isolation in primary motor cortex cells, J. Neurosci. Meth. 246 (2015) 106–118. Crossref, Medline, Web of Science, Google Scholar
- 36. , Regularizing common spatial patterns to improve BCI designs: Unified theory and new algorithms, IEEE Trans. Biomed. Eng. 58(2) (2011) 355–362. Crossref, Medline, Web of Science, Google Scholar
- 37. , Local temporal common spatial patterns for robust single-trial EEG classification, IEEE Trans. Neural Syst. Rehab. Eng. 16(2) (2008) 131–139. Crossref, Medline, Web of Science, Google Scholar
- 38. , Local temporal correlation common spatial patterns for single trial EEG classification during motor imagery, Comput. Math. Meth. Med. (2013) 591216. Medline, Web of Science, Google Scholar
- 39. , Robustifying EEG data analysis by removing outliers, Choas Complexity Lett. 2(2) (2009) 251–266. Google Scholar
- 40. , Integrating EEG and MEG signals to improve motor imagery classification in brain–computer interface, Int. J. Neural Syst. 29(1) (2019) 1850014. Link, Web of Science, Google Scholar
- 41. , An auditory-tactile visual saccade-independent p300 brain–computer interface, Int. J. Neural Syst. 26(1) (2016) 1650001. Link, Web of Science, Google Scholar
- 42. , Multivariate estimation with high breakdown point, Mathematical statistics and applications 8(283–297) (1985) 37. Google Scholar
- 43. , Minimum Covariance Determinant and Extensions (Wiley, 2017), p. e:1421. Google Scholar
- 44. , A fast algorithm for the minimum covariance determinant estimator, Technometrics 41(3) (1999) 212–223. Crossref, Web of Science, Google Scholar
- 45. , Robust monitoring of industrial processes in the presence of outliers in training data, Ind. Eng. Chem. Res. 57(24) (2018) 8230–8239. Crossref, Web of Science, Google Scholar
- 46. , Robust common spatial patterns for EEG signal preprocessing, 2008 30th Annual Int. Conf. IEEE Eng. Med. Biol. Soc.
20–25 August 2008 ,Vancouver, Canada , pp. 2087–2090. Google Scholar - 47. , Separable common spatio-spectral patterns for motor imagery BCI systems, IEEE Trans. Biomed. Eng. 63(1) (2016) 15–29. Crossref, Medline, Web of Science, Google Scholar
- 48. , Time-series discrimination using feature relevance analysis in motor imagery classification, Neurocomputing 151 (2015) 122–129. Crossref, Web of Science, Google Scholar
- 49. , An efficient hybrid linear and kernel CSP approach for EEG feature extraction, Neurocomputing 73(1–3) (2009) 432–437. Crossref, Web of Science, Google Scholar
- 50. , Adaptive semi-supervised classification to reduce intersession non-stationarity in multiclass motor imagery-based brain-computer interfaces, Neurocomputing 159 (2015) 186–196. Crossref, Web of Science, Google Scholar
- 51. , Correlation-based channel selection and regularized feature optimization for MI-based BCI, Neural Netw. 118 (2019) 262–270. Crossref, Medline, Web of Science, Google Scholar
- 52. , Bispectrum-based channel selection for motor imagery based brain–computer interfacing, IEEE Trans. Neural Syst. Rehab. Eng. 28(10) (2020) 2153–2163. Crossref, Medline, Web of Science, Google Scholar
- 53. , Least median of squares regression, J. Am. Stat. Assoc. 79(388) (1984) 871–880. Crossref, Web of Science, Google Scholar
- 54. , High-breakdown robust multivariate methods, Stat. Sci. 23(1) (2008) 92–119. Crossref, Web of Science, Google Scholar
- 55. ,
Combining SVMs with various feature selection strategies , in Feature Extraction: Foundations and Applications, (Springer, Berlin, 2006), pp. 315–324. Crossref, Google Scholar - 56. Q. Gu, Z. Li and J. Han, Generalized fisher score for feature selection, preprint (2012), arXiv:1202.3725. Google Scholar
- 57. , An overview of statistical learning theory, IEEE Trans. Neural Netw. 10(5) (1999) 988–999. Crossref, Medline, Web of Science, Google Scholar
- 58. , Boosting bit rates in noninvasive EEG single-trial classifications by feature combination and multiclass paradigms, IEEE Trans. Biomed. Eng. 51(6) (2004) 993–1002. Crossref, Medline, Web of Science, Google Scholar
- 59. , The non-invasive Berlin brain–computer interface: Fast acquisition of effective performance in untrained subjects, NeuroImage 37(2) (2007) 539–550. Crossref, Medline, Web of Science, Google Scholar
- 60. , Composite common spatial pattern for subject-to-subject transfer, IEEE Signal Process. Lett. 16(8) (2009) 683–686. Crossref, Web of Science, Google Scholar
- 61. , Improvement of classification accuracy in a phase-tagged steady-state visual evoked potential-based brain computer interface using multiclass support vector machine, Biomed. Eng. Online 12 (2013) 46. Crossref, Medline, Web of Science, Google Scholar
- 62. , A comparison of classification techniques for the P300 speller, J. Neural Eng. 3(4) (2006) 299–305. Crossref, Medline, Web of Science, Google Scholar
- 63. , Brain–computer interface after nervous system injury, Neuroscientist 20(6) (2014) 639–651. Crossref, Medline, Web of Science, Google Scholar
- 64. , Tangent space features-based transfer learning classification model for two-class motor imagery brain–computer interface, Int. J. Neural Syst. 29(10) (2019) 1950025. Link, Web of Science, Google Scholar
- 65. , Comparing recalibration strategies for electroencephalography-based decoders of movement intention in neurological patients with motor disability, Int. J. Neural Syst. 28(7) (2018) 1750060. Link, Web of Science, Google Scholar
- 66. , Robust filter bank common spatial pattern (RFBCSP) in motor-imagery-based brain–computer interface, 2009 Annual Int. Conf. IEEE Eng. Med. Biol. Soc.
3–6 September 2009 ,Minneapolis, MN, USA , pp. 578–581. Google Scholar - 67. , A new neural dynamic classification algorithm, IEEE Trans. Neural Netw. Learn. Syst. 28(11) (2017) 3074–3083. Crossref, Medline, Web of Science, Google Scholar
- 68. , Enhanced probabilistic neural network with local decision circles: A robust classifier, Integr. Comput.-Aided Eng. 17(3) (2010) 197–210. Crossref, Web of Science, Google Scholar
- 69. , A dynamic ensemble learning algorithm for neural networks, Neural Comput. Appl. 32(11) (2020) 8675–8690. Crossref, Web of Science, Google Scholar
- 70. , FEMa: A finite element machine for fast learning, Neural Comput. Appl. 32(10) (2020) 6393–6404. Crossref, Web of Science, Google Scholar
Remember to check out the Most Cited Articles! |
---|
Check out our titles in neural networks today! |