Atrial fibrillatory signal estimation using blind source extraction algorithm based on high-order statistics

The analysis and the characterization of atrial fibrillation (AF) requires, in a previous key step, the extraction of the atrial activity (AA) free from 12-lead electrocardiogram (ECG). This contribution proposes a novel non-invasive approach for the AA estimation in AF episodes. The method is based on blind source extraction (BSE) using high order statistics (HOS). The validity and performance of this algorithm are confirmed by extensive computer simulations and experiments on real-world data. In contrast to blind source separation (BSS) methods, BSE only extract one desired signal, and it is easy for the machine to judge whether the extracted signal is AA source by calculating its spectrum concentration, while it is hard for the machine using BSS method to judge which one of the separated twelve signals is AA source. Therefore, the proposed method is expected to have great potential in clinical monitoring.

[1]  José Millet-Roig,et al.  Atrial activity extraction for atrial fibrillation analysis using blind source separation , 2004, IEEE Transactions on Biomedical Engineering.

[2]  Andrzej Cichocki,et al.  Adaptive blind signal and image processing , 2002 .

[3]  Zhang Yi,et al.  Extraction of temporally correlated sources with its application to non-invasive fetal electrocardiogram extraction , 2006, Neurocomputing.

[4]  Wei Lu,et al.  ICA with Reference , 2006, Neurocomputing.

[5]  Leif Sörnmo,et al.  Spatiotemporal QRST cancellation techniques for analysis of atrial fibrillation , 2001, IEEE Transactions on Biomedical Engineering.

[6]  Pablo Laguna,et al.  Vectorcardiographic loop alignment and the measurement of morphologic beat-to-beat variability in noisy signals , 2000, IEEE Transactions on Biomedical Engineering.

[7]  Philip Langley,et al.  Comparison of atrial signal extraction algorithms in 12-lead ECGs with atrial fibrillation , 2006, IEEE Transactions on Biomedical Engineering.

[8]  Jimin Ye,et al.  Blind source separation with unknown and dynamically changing number of source signals , 2006, Science in China Series F: Information Sciences.

[9]  José Millet-Roig,et al.  Atrial activity extraction from atrial fibrillation episodes based on maximum likelihood source separation , 2005, Signal Process..

[10]  Shun-ichi Amari,et al.  Sequential blind signal extraction in order specified by stochastic properties , 1997 .

[11]  Zhang Yi,et al.  Robust extraction of specific signals with temporal structure , 2006, Neurocomputing.

[12]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[13]  Wei Lu,et al.  Approach and applications of constrained ICA , 2005, IEEE Transactions on Neural Networks.

[14]  Gang Wang,et al.  An Extended Online Fast-ICA Algorithm , 2006, ISNN.

[15]  César Sánchez,et al.  Derivation of Atrial Surface Reentries Applying ICA to the Standard Electrocardiogram of Patients in Postoperative Atrial Fibrillation , 2006, ICA.

[16]  P. Langley,et al.  Frequency analysis of atrial fibrillation , 2000, Computers in Cardiology 2000. Vol.27 (Cat. 00CH37163).

[17]  José Millet-Roig,et al.  Spatiotemporal blind source separation approach to atrial activity estimation in atrial tachyarrhythmias , 2005, IEEE Transactions on Biomedical Engineering.

[18]  Wei Liu,et al.  A class of novel blind source extraction algorithms based on a linear predictor , 2005, 2005 IEEE International Symposium on Circuits and Systems.

[19]  Allan Kardec Barros,et al.  Extraction of Specific Signals with Temporal Structure , 2001, Neural Computation.

[20]  Zhang Yi,et al.  Extraction of a source signal whose kurtosis value lies in a specific range , 2006, Neurocomputing.