||By using a force platform and a 12-channel EEG measuring device, this work investigates the correlation between EEG activities and posture balance features. |
Fifteen young and healthy people agreed to quietly stand on a force platform with their eyes open and closed. Initially, with their arms relaxed on the side of the body and eyes open, participants were asked to look straight ahead ﬁxating their gaze at a black spot positioned at eye level at a distance of approximately 2 m. This period lasts for 10 min. Next, after 5 min of rest, the same participant was instructed to repeat the same quiet standing task with eyes closed for 10 min. The frequency response of the EEG was computed by the Welch’s method. To eliminate the artifacts, independent component analysis and an outlier method have been employed to suppress the influences of noise and eliminate contaminated signal intervals.
This work computed eight force platform features to characterize the posture stability. Six of these eight stability features were derived from the trajectory of the center of pressure (COP). These six features are the sway area, mean distance, mean velocity, mean velocity in the anterior-posterior direction, mean velocity in the medial-lateral direction, mean frequency of the COP. Based on the ground reaction force (GRF), this study also proposes two additional features to assess the posture stability.
This work investigates four types of correlations between the EEG spectra and stability features. The first (eyes open correlation) and second (eyes closed correlation) were computed directly from the eyes open and eyes closed conditions, respectively. The third (variation correlation) computes the correlation between the eyes open and eyes closed difference of the EEG response and the eyes open and eyes closed differences of the stability features. The fourth (variation ratio correlation) calculate the eyes open and eyes closed difference divided by eyes open, and computes the correlation between the EEG response and stability features.
The results can be summarized as follows. First, negative variation correlations between the EEG energy of the 4-30 Hz band and the COP mean velocity features were identified in several EEG channels. Second, similar finding also occur on the α band of the EEG signal. Third, for the eyes open and eyes closed correlations, the COP mean velocity features are positively correlated with the 14~18Hz frequency band energy for several EEG channels.