A comparison of subject-dependent and subject-independent channel selection strategies for single-trial P300 brain computer interfaces
- 52 Downloads
Brain computer interfaces (BCI) represent an alternative for patients whose cognitive functions are preserved, but are unable to communicate via conventional means. A commonly used BCI paradigm is based on the detection of event-related potentials, particularly the P300, immersed in the electroencephalogram (EEG). In order to transfer laboratory-tested BCIs into systems that can be used by at homes, it is relevant to investigate if it is possible to select a limited set of EEG channels that work for most subjects and across different sessions without a significant decrease in performance. In this work, two strategies for channel selection for a single-trial P300 brain computer interface were evaluated and compared. The first strategy was tailored specifically for each subject, whereas the second strategy aimed at finding a subject-independent set of channels. In both strategies, genetic algorithms (GAs) and recursive feature elimination algorithms were used. The classification stage was performed using a linear discriminant. A dataset of EEG recordings from 18 healthy subjects was used test the proposed configurations. Performance indexes were calculated to evaluate the system. Results showed that a fixed subset of four subject-independent EEG channels selected using GA provided the best compromise between BCI setup and single-trial system performance.
KeywordsSubject-independent channel selection Brain-computer interface P300 Genetic algorithm Recursive feature elimination
This work was financially supported by the National University of Entre Ríos (UNER), PID No. 6163.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 1.Wolpaw JR, Birbaumer N, Heetderks WJ, McFarland D, Peckham PH, Schalk G, Donchin E, Quatrano LA, Robinson CJ, Vaughan TM (2000) Brain-computer interface technology: a review of the first international meeting. IEEE Trans Rehabil Eng 8:164–173. https://doi.org/10.1109/TRE.2000.847807 CrossRefPubMedGoogle Scholar
- 6.Acevedo R, Atum Y, Gareis I, Biurrun Manresa J, Medina Bañuelos V, Rufiner L (2018) A comparison of feature extraction strategies using wavelet dictionaries and feature selection methods for single trial P300-based BCI. Med Biol Eng Comput 57:589–600. https://doi.org/10.1007/s11517-018-1898-9 CrossRefPubMedGoogle Scholar
- 10.Shih EI, Shoeb AH, Guttag JV (2009) Sensor selection for energy-efficient ambulatory medical monitoring. Proc 7th Int Conf Mob Syst Appl Serv - Mobisys ’09 347. https://doi.org/10.1145/1555816.1555851
- 15.Li Y, Wang L-Q, Hu Y (2009) Localizing P300 generators in high-density event- related potential with fMRI. Med Sci Monit 15:47–53Google Scholar
- 25.Ledesma-Ramírez C, Bojorges-Valdéz E, Yáñez-Suárez O, et al (2010) An open-access P300 Speller Database. In: Fourth International Brain-Computer Interface Meeting. pp 3–4Google Scholar
- 27.Kee CY, Kuppan Chetty RM, Khoo BH, Ponnambalam SG (2012) Genetic algorithm and Bayesian linear discriminant analysis based channel selection method for P300 BCI. In: Communications in Computer and Information Science. pp 226–235Google Scholar
- 29.Hasan IH, Ramli AR, Ahmad SA (2014) Utilization of genetic algorithm for optimal EEG channel selection in brain-computer interface application. 2014 4th Int Conf Artif Intell with Appl Eng Technol 97–102. https://doi.org/10.1109/ICAIET.2014.25
- 30.Yang J, Singh H, Hines EL, Schlaghecken F, Iliescu DD, Leeson MS, Stocks NG (2012) Channel selection and classification of electroencephalogram signals: An artificial neural network and genetic algorithm-based approach. Artif Intell Med 55:117–126. https://doi.org/10.1016/j.artmed.2012.02.001 CrossRefPubMedGoogle Scholar
- 32.Atum YV, Biurrun Manresa JA, Rufiner L, Acevedo RC (2015) Genetic feature selection for a P300 brain computer interface. In: IFMBE ProceedingsGoogle Scholar
- 35.Van Dijck G, Van Hulle MM, Wevers M (2004) Genetic algorithm for feature subset selection with exploitation of feature correlations from continuous wavelet transform: a real-case application. Int Conf Comput Intell 1:34–38Google Scholar
- 36.Atum Y, Gareis I, Gentiletti G, et al (2010) Genetic feature selection to optimally detect P300 in Brain Computer Interfaces. 2010 Annu Int Conf IEEE Eng Med Biol Soc EMBC’10 3289–3292. https://doi.org/10.1109/IEMBS.2010.5627254
- 37.Pacheco M, Atum Y, Acevedo R, Rufiner L (2016) Evaluation of different parents selection methods in a genetic algorithm wrapper for P300 BCI. In: XXV Brazilian Congress of Biomedical Engineering (CBEB 2016). pp 1433–1436Google Scholar
- 38.Duda RO, Hart PE, Stork DG (2001) Pattern Classification. John Wiley, New York Sect 680Google Scholar
- 42.Gonzalez A, Nambu I, Hokari H, et al (2013) Towards the classification of single-trial event-related potentials using adapted wavelets and particle swarm optimization. Proc - 2013 IEEE Int Conf Syst Man, Cybern SMC 2013, pp 3089–3094. https://doi.org/10.1109/SMC.2013.527
- 43.Shahriari Y, Erfanian A (2011) A mutual information based channel selection scheme for P300-based brain computer interface. 5th Int IEEE/EMBS Conf Neural Eng 434–437. https://doi.org/10.1109/NER.2011.5910579