Motor imagery

Motor Imagery based BMI design to control a power wheelchair is presented. Synchronous and asynchronous experimental results are presented and discussed. Real-time... 

G. Pfurtscheller and C. Neuper, Motor Imagery and Direct Brain-Computer Communication, IEEE Transactions Vol. 89, No.7, pp 1123-1134,1... 

S. Cososchi, R. Stungaru, A Unureanu and M. Ungureanu, EEG Features Extraction for Motor Imagery, Proc. 2S IEEE EMBS Annual Inti. Conf, USA, pp 1142 - 1145, 2006. 

Hema C.R., Paulraj M.P., S.Yaacob, A.H. Adorn, Nagarajan R, Motor Imagery Signal Classification for a Four State Brain Machine Interface , International Journal of Biomedical Sciences, Vol.3 No.l, pp. 76-81 December 2008. 

Comparative Analysis of the Optimal Performance Evaluation for Motor Imagery... 

Keywords— Comparative analysis. Optimal performance. Motor imagery, EEG, Brain computer interface. 

For an accuracy evaluation of the proposed algorithm, a BCI experiment was performed for 10 subjects. The 10 subjects in the second set of experiments were selected from the first experiments. They understood the experimental methods and process as they had obtained some measure of BCI experience. The mental task was motor imagery for the left or right arm. The experimental protocol consisted of a... 

The experiment s subjects had no mental illness and were on no medications at the time of the study. The experimental environment was isolated by a curtain, and the subjects concentrated on the monitor and the speaker to detect the visual and auditory cues. Subjects did not move during the experiment. The experiment s manager monitored the subject s activity, and environmental changes. The motor imagery of the subjects was the imagination of moving the left arm or the right arm [3]. 

The effect of the performance factor is important in evaluations of the optimal performance of motor-imagery-based EEG-BCI algorithms. There are five types of performance factors, as shown in Table 1, the EEG mode, feature calculation, selected CSP channel number, selected classifier and the window size. The EEG mode has two selections, ERD and RP. Feature calculation has the four selections of log (var) mean, sum (abs) and log (power). The CSP channel number has the four selections of 2, 4, 8... 

Cross Evaluation for Characteristics of Motor Imagery Using Neuro-feedback... 

The purpose of this paper is to evaluate the performance of EEG-BCI as a motor imagery characteristic for use as a ubiquitous health service to determine of optimal evaluation performance conditions for a range of subjects. The paper also seeks to determine the average conditions under which maximum performance can be attained with a view to proposing the evaluation technology for use with EEG-BCI-based medical instruments while guaranteeing the safety and confidence of such devices. 

Pfurtscheller G., C. Neuper, Motor imagery activates primary sensorimotor area in humans , Neuroscience Letters, vol. 239, Issue 2-3, Dec. 1997, pp. 65-68... 

Wei-Yen Hsu, Yung-Nien Sim (2009) EEG-based motor imagery analysis using weighted wavelet transform features. Journal of Neuroscience Methods 176 310-318... 

Evaluation of Motor Imagery Using Combined Cue Based EEG-Brain Computer... 

This experiment evaluates the effects of the combined cue. There were six types of combinations between the training cues and testing cues. Each experiment had 50 trials, and each trial had a two-second blank screen, a two-second fixation cross, and a four-second visual or auditory cue. Blank screens represented the interval of each trial, and the fixation cross denoted the preparation of motor imagery. 

Generally, a BCI experiment uses visual cues and auditory cues. A visual cue tjipicaUy is a graphical symbol and/or directional signs for motor imagery. Auditory cues use sound effects instead of a visual cue. These methods evaluate the brain s information process performance. In a preliminary study, the classification accuracy was tested using visual cues and auditory cues. In this study, visual cues and auditory cues were combined and simultaneously presented. Thus, the subject trained using a combined cue. This method tested the performance between an individual cue and a combined cue [3]. 

Sitaram R., Zhang H., et al., (2007) Temporal classification of multichannel near-infrared spectroscopy signals of motor imagery for developing a hrain-computer interface. Neuroimage 34 1416 1427. 

---