Abstract— In this paper we propose to apply spatio-temporal filtering to fetal ECG extraction from the multichannel maternal abdominal bioelectric signals. The method performs weighted addition of the assumed number of time samples of the respective measured signal channels. The filter weights are calculated during the learning phase of the method. After this learning phase, the filter operation is very simple, similar to usual filtering, nevertheless extraordinarily effective. Our experiments show that it can simultaneously accomplish maternal ECG suppression and fetal QRS complexes enhancement for their detection. The detection tests are performed to show the method influence on this crucial operation.
Index Terms— Fetal electrocardiogram, maternal ECG suppression, spatio-temporal filtering, generalized eigendecomposition.
M. Kotas is with the Division of Biomedical Electronics, Institute of Electronics, Silesian University of Technology, Gliwice, Poland (e-mail: marian.kotas@polsl.pl).
J. Błaszczyk and T. Moroń are with the Faculty of Automatic Control Electronics and Informatics, Silesian University of Technology, Gliwice, Poland (e-mail: jakubla466@student.polsl.pl, tomasz.moron@polsl.pl).
[PDF]
Cite: Marian Kotas, Jakub Błaszczyk, and Tomasz Moroń, " Spatio-Temporal FIR Filter for Fetal ECG Extraction," International Journal of Information and Electronics Engineering vol. 5, no. 1, pp. 10-14, 2015.
