ICA and Wavelet Transform to Separate Atrial Fibrillation Wave
Wavelet Transform (WT) and Independent Component Analysis (ICA) were addressed to study separating atrial fibrillation (AF) from body surface electrocardiosignal (ECS). The subjects are in supine, ECS from the lead called V1 have been recorded continuously for 8 minutes and sampled at 500Hz. Wavelet Transform decomposed ECS into two detailed levels and one approximation. Wavelet Transform also rebuilt the signal from the approximation. ICA was employed to decompose the original signal and the signal reconstructed by WT into three groups. The result expressed the approach can separate independent AF wave and cardiac ventricular wave. The separating wave can help us to diagnose in clinic. Comparing these two similar results, it suggests that all valuable components have been reserved after doing Wavelet Transform. Combining the techniques of Wavelet, ICA and other signal processing methods will be the tendency of complex physiological signal processing.
Zhangyong Li Shengrong Liu Wei Wang Zhonghua He Jinmin Wei Zhengxiang Xie
College of Bioinformation Chongqing University of Posts and Telecommunications Chongqing 400065 China
国际会议
南宁
英文
2007-07-20（万方平台首次上网日期，不代表论文的发表时间）