Determining fetal heart rate using independent component analysis

Abstract

Recording the fetal ECG (fECG) is a basic method of assessing the condition of the heart of the fetus. In most cases this signal is collected in a non-invasive manner, thus it is corrupted by the mother's ECG (mECG). The separation of these two signals is still a challenge. On one hand, there is no standard for the placement of abdominal ECG electrodes and on the other hand, there exists no widely accepted general separation method yet. Our proposed approach is based on applying blind source separation via the FastICA method to baseline-free ECG signals. This method will offer the position of the R peaks of the fECG making it possible to determine the heart rate of the fetus. We found that the outcome of this method is highly dependent of the quality of the measurements and the amplitude of the fECG relative to the mECG.

An Efficient Approach for Fetal ECG Extraction

“…The FECG signals are often procured by 2 procedures, specifically fetal scalp electrode and maternal abdomen skin electrode [1]. Different methods have been forth put for fetal ECG extrication for instance matching pursuits [2] , Independent component analysis(ICA) [3][4][5],blind source extraction(BSE) [6],adaptive filtering [7], support vector machine [8], singular value decomposition(SVD) [9], wavelet analysis [10], extended Kalman filtering [11], auto-correlation and cross-correlation techniques [12]. Adaptive filtering combined with neural network is a powerful scheme for FECG extraction.…”