Non-Adaptive Methods of Fetal ECG Signal Processing

• Main Authors: Radana Kahankova, Rene Jaros, Radek Martinek, Janusz Jezewski, He Wen, Michal Jezewski, Aleksandra Kawala-Janik
• Format: Article
• Language: English
• Published: VSB-Technical University of Ostrava 2017-01-01
• Series: Advances in Electrical and Electronic Engineering
• Subjects: blind source separation; ecg extraction, fetal electrocardiogram (ecg); independent component analysis; non-adaptive filtration; non-invasive fetal monitoring; principal component analysis.
• Online Access: http://advances.utc.sk/index.php/AEEE/article/view/2196
• ISSN: 1336-1376; 1804-3119

Abdominal fetal ElectroCardioGrams (fECGs) carry a wealth of information about the fetus including fetal Heart Rate (fHR) and signal morphology during different stages of pregnancy. Here we report our results on the implementation and evaluation of two non-adaptive signal processing methods suitable for fECG signal extraction, namely: the Independent Component Analysis (ICA) and the Principal Component Analysis (PCA) Methods. We used the fetal heart rate extracted from fECG signals (in Beats Per Minute - BPM) and Signal-to-Noise Ratio (SNR) as effective performance evaluation metrics for our applied methods. Our findings demonstrated that given adequate SNR, these methods produced excellent results in accurate determination of fHR. Furthermore, we found out that compared to the PCA Method, the ICA Method produces a lower variance in the detection of the fHR.