Fetal heart rate detection and monitoring from noninvasive abdominal ECG recordings

Along the pregnancy, fetal heart rate (FHR) can be monitored noninvasively using Doppler ultrasound. Unfortunately, this technique is inaccurate and provides a relatively low positive predictive value. Performing a fetal electrocardiogram (FECG) waveform analysis it can be obtained useful information about physiological state of the fetus. FECG can be extracted from abdominal electrocardiogram (aECG) recordings by means of multiple electrodes placed on the abdomen of the mother. The main research goal of the present paper is to extract fECG signals from a collection of 4 channel aECG preprocessed recordings using Fast Independent Component Analysis (ICA) algorithm and to denoise extracted fECG signals by post-processing using Undecimated Wavelet Transform (UWT) and a Fast Fourier Transform (FFT) - Inverse Fast Fourier Transform (IFFT) pair. The degree of fetal bradycardia during labor (depression of mean FHR below 110 BPM at full term) was established for the three patients investigated, after a registration of BPM variations at various time intervals for all 5 minutes of recordings.