北京生物醫(yī)學(xué)工程

基于典型相關(guān)分析和小波變換的眼電偽跡去除

Automatic Removal of Ocular Artifacts in EEG Signals by Using CCA and Wavelet Transformation

作者：趙春煜邱天爽

單位：大連理工大學(xué)電子信息與電氣工程學(xué)部（大連116024）

分類號：

出版年·卷·期（頁碼）：2011·30·5（474-479）

摘要：

目的針對腦電信號中眼電偽跡去除尚存在的問題，提出一種基于典型相關(guān)分析與小波變換的
(wavelet-enhanced canonical correlation analysis, wCCA)自動去除眼電偽跡的算法。方法首先，
充分利用腦電信號和眼電偽跡的空間分布特征，將基于典型相關(guān)分析的盲源分離算法分別應(yīng)用于左右腦
區(qū)的混合信號中，從而保證典型相關(guān)分析分解得到的第一個(gè)典型相關(guān)變量（即左右腦區(qū)之間的最公共成
分），就是眼電偽跡分量。然后為了恢復(fù)泄漏在該偽跡分量中的腦電成分，對偽跡分量進(jìn)行小波閾值濾
波，將高于某一閾值的小波系數(shù)置零，而保留低于閾值的系數(shù)。結(jié)果與其他三種基于盲源分離去除眼電
偽跡的方法相比較，該方法在有效地自動去除眼電偽跡的同時(shí)，很好地保留了潛在的腦電信號，去除效
果明顯優(yōu)于其他三種方法。結(jié)論由于該算法簡單，處理速度較快，因此應(yīng)用于實(shí)時(shí)的腦機(jī)接口系統(tǒng)中更
具優(yōu)越性，為后續(xù)腦電信號的特征提取和分類分析提供了良好的基礎(chǔ)。

Objective A new method of ocular artifacts removal in EEG
(electroencephalography) recordings, wavelet-enhanced canonical correlation analysis
(wCCA), is presented in this paper. Methods Firstly, considering the differences between
the spatial distributions of the EEG signals and the EOG signals, CCA is applied to the
mixed signals of left and right brain separately. There is no need to identify the artifact
component by subjective visual inspection, because the first canonical component found by
CCA for each dataset, also the most common component between the left and right hemisphere,
is definitely related to artifacts. Then wavelet thresholding is employed to recover the
cerebral activities leaked into this artifact component. The performance of the proposed
method is compared to the three popular ocular artifacts removal methods CCA, second-order
-blind identification(SOBI) and wavelet independent component analysis(ICA), in terms of
correlation coefficient and signal-to-artifact ratio (SAR). Results It shows that wCCA’s
performance is better than those of the other three methods for removing the most ocular
artifacts from EEG recording automatically without altering the cerebral components.
Conclusions Since wCCA is simple and rapid, it is more advantageous to be applied in true
time brain-computer interface system than the other three, and provides a good groundwork
for the feature extraction and classification analysis of electroencephalography.

參考文獻(xiàn)：

［1］Romero S, Mananas MA, Barbanoj MJ. Ocular Reduction in EEG Signals Based on Adaptive
Filtering, Regression and Blind Source Separation ［J］. Annals of Biomed Engin, 2009, 37
(1): 176-191.
［2］Congedo M, Pailler GC, Jutten. On the blind source separation of human
electroencephalogram by approximate joint diagonalization of second order statistics ［J］.
Clin Neurophy, 2008, 119(12)： 2677-2686.
［3］Makarov VA, Castellanos NP. Recovering EEG brain signals: Artifact suppression with
wavelet enhanced independent component analysis ［J］. Neuro Meth, 2006, 158(2)： 300-312.
［4］Inuso G, La Foresta F, Mammone N, et al. Wavelet-ICA methodology for efficient
artifact removal from Electroencephalographic recordings ［C］//International Joint
Conference on Neural Networks. Orlando: IEEE, 2007:1524-1529.
［5］Delorme A, Sejnowski T, Makeig S. Enhanced detection of artifacts in EEG data using
higher-order statistics and independent component analysis ［J］. of Neurosci Meth, 2007,
34(4): 1443-1449.
［6］高莉, 黃力宇. 基于獨(dú)立分量分析的自適應(yīng)最大熵算法對腦電干擾的識別與剔除［J］. 航天醫(yī)學(xué)
與醫(yī)學(xué)工程, 2008, 21(2): 142-146.
［7］Clercq WD, Vergult A, Vanrumste B, et al. Canonical correlation analysis applied to
remove muscle artifacts from the electroencephalogram ［J］. IEEE Trans Biomed Eng, 2006,
53(12): 2583-2587.
［8］張莉, 何傳紅, 何為. 典型相關(guān)分析去除腦電信號中的眼電偽跡的研究［J］. 計(jì)算機(jī)工程與應(yīng)用
, 2009, 45(31): 218-220.
［9］Siew-Cheok NG, Raveendran P. Enhanced u rhythm extraction using blind source
separation and wavelet transform ［J］. IEEE Trans Biomed Eng, 2009, 56(8): 2024-2034.
［10］Romero S, Mananas MA, Barbanoj MJ. A comparative study of automatic techniques for
ocular artifact reduction in spontaneous EEG signals based on clinical target variables: A
simulation case ［J］. Comp in Biolog and Med, 2008, 38(3): 348-360.

服務(wù)與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請登錄！點(diǎn)此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频