Improved mECG Removal and fECG Extraction by Integrated Periodic Components Analysis and Singular Value Decomposition

Stillbirth remains a global concern, mainly arising from complications affecting over 20% of pregnancies. Early detection and diagnosis of these complications are crucial for timely medical intervention and improved patient outcomes. Fetal electrocardiography (fECG) emerges as a promising tool for non-invasive and comprehensive monitoring of fetal well-being. However, extracting a reliable fECG signal from the electrophysiological signals obtained with electrodes positioned on the maternal abdomen is challenging due to its reduced amplitude and the presence of various interference, primarily the maternal ECG (mECG). This work proposes an improved denoising approach based on the iterative combination of Periodic Component Analysis (πCA) and Singular Value Decomposition (SVD) to remove the mECG interference and enhance the accurate extraction of the fECG signals from multichannel electrophysiological signals. In-silico, the proposed approach has shown superior performance compared to existing methods in the literature, demonstrating significant enhancements across various noise levels. An in-vivo analysis validates the capability of the proposed method to provide a reliable fECG, ensuring accurate extraction of clinically relevant parameters. The outcomes of this work hold promise for advancing the diagnosis of pregnancy complications and fetal pathologies, promoting better monitoring of fetal well-being and timely intervention in case of adverse events.