What ICA Provides for ECG Processing: Application to Noninvasive Fetal ECG Extraction

Reza Sameni (1, 2) , Christian Jutten (2) , M.B. Shamsollahi (1)

1 BiSIPL - Biomedical Signal and Image Processing Laboratory [Teheran]
2 LIS - Laboratoire des images et des signaux

In recent studies, Independent Component Analysis (ICA) has been used for the analysis of multi-channel ECG recordings. However most of these works have been carried out from the signal processing perspective. In this work, the single dipole vector theory of the heart and the ECG dimensionality are studied from the source separation viewpoint. Based on this study, the interpretation of the components extracted from multi-channel ECG and maternal abdominal recordings, and their relationship with the vectorcardiogram representation of the cardiac dipole are presented. The results of this study can be used for the extraction of meaningful clinical indexes, based on ICA techniques.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Reza Sameni, Christian Jutten, M.B. Shamsollahi. What ICA Provides for ECG Processing: Application to Noninvasive Fetal ECG Extraction. Proc. of the International Symposium on Signal Processing and Information Technology (ISSPIT'06), Vancouver, Canada, Aug. 27-30, 2006, Aug 2006, Vancouver, Canada. pp.656-661. ⟨hal-00174338⟩