Browsing by Author "Song, Meng-Shue"

Now showing 1 - 2 of 2

Developing a Few-channel Hybrid BCI System by Using Motor Imagery with SSVEP Assist
(2014)
Ko, Li-Wei
;
Lin, Shih-Chuan
;
Song, Meng-Shue
;
Komarov, Oleksii
;
生物資訊及系統生物研究所
Generally, Steady-State Visually Evoked Potentials (SSVEP) has widely recognized advantages, like being easy to use, requiring little user training [1], while Motor Imagery (MI) is not easy to introduce for some subjects. This work introduces a hybrid brain-computer interface (BCI) combines MI and SSVEP strategies - such an approach allows us to improve performance and universality of the system, and also the number of EEG electrodes from 32 to 3 in central area can increase the efficiency of EEG preprocessing to design an effective and easy way to use hybrid BCI system. In this study the Common Spatial Pattern (CSP) algorithm was introduced as a feature extraction method, which provides a high accuracy in event-related synchronization/desynchronization (ERS/ERD)-based BCI. The four most common classifiers (KNNC, PARZENDC, LDC, SVC) were used for accuracy estimation. Results show that support vector classifier (SVC) and K-nearest-neighbor (KNN) classifier provide better performance than others, and it is possible to reach the same good accuracy using 3-channel (C3, Cz, C4) hybrid BCI system, as with usual 32-channel system .
Development of SSVEP-based BCI using Common Frequency Pattern to Enhance System Performance
(2014)
Ko, Li-Wei
;
Lin, Shih-Chuan
;
Liang, Wei-Gang
;
Komarov, Oleksii
;
Song, Meng-Shue
Brain Computer Interface(BCI) systems provide an additional way for people to interact with external environment without using peripheral nerves or muscles[1]. In a variety of BCI systems, a BCI system based on the steady-state visual evoked potentials (SSVEP) is one most common system known for application, because of its ease of use and good performance with little user training. In this study, the common frequency pattern method (CFP) is used to improve the accuracy of our EEG-based SSVEP BCI system. There are four basic classifiers (SVM, KNNC, PARZENDC, LDC) in this paper to estimate the accuracy of our SSVEP system. Without using CFP, the highest accuracy of the EEG-based SSVEP system was 80%. By using CFP, the accuracy could be upgraded to 95%.