Combining CCA and CFP for enhancing the performance in the hybrid BCI system

Li-Wei Ko, S. Sai Kalyan Ranga

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Hybrid Brain Computer Interface (BCI) is gaining attention as it can provide better performance or increase the number of user commands to control an external device. Hybrid BCI system using Motor imagery (MI) and Steady-state visually evoked potential (SSVEP) is one such system. Maintaining the performance during channel reduction is important in practical applications. In this paper we propose a combined feature extraction method using Canonical Correlation Analysis (CCA) and Common Frequency Pattern (CFP) method, where the features obtained from these methods were combined for classification. We used LDC and PARZEN for estimating the classification accuracy for the proposed method and individual method. Highest accuracy of 96.1 % is obtained for combined feature method (CCA+CFP). Whereas, the accuracy is 89.6% with CCA and 91.6% with CFP method. A significance test has shown that the performance of the proposed method is significantly different from both the individual methods (p < 0.05).

Original languageEnglish
Title of host publicationProceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages103-108
Number of pages6
ISBN (Electronic)9781479975600
DOIs
StatePublished - 1 Jan 2015
EventIEEE Symposium Series on Computational Intelligence, SSCI 2015 - Cape Town, South Africa
Duration: 8 Dec 201510 Dec 2015

Publication series

NameProceedings - 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015

Conference

ConferenceIEEE Symposium Series on Computational Intelligence, SSCI 2015
CountrySouth Africa
CityCape Town
Period8/12/1510/12/15

Fingerprint Dive into the research topics of 'Combining CCA and CFP for enhancing the performance in the hybrid BCI system'. Together they form a unique fingerprint.

Cite this