Joint Multiple Subspace-based BSS Method for Fetal Heart Rate Extraction from Non-invasive Recordings

Area of Science:

• Biomedical Engineering
• Signal Processing
• Maternal-Fetal Medicine

Background:

• Non-invasive fetal electrocardiography (fECG) is crucial for fetal disease detection but is often hindered by maternal ECG interference and motion artifacts.
• Existing methods struggle to reliably isolate fetal heart rate (HR) signals due to severe signal contamination.

Purpose of the Study:

• To develop and validate a novel joint multiple subspace-based blind source separation (BSS) approach for accurate fetal heart rate (HR) extraction.
• To significantly reduce the impact of maternal electrocardiograph (ECG) and motion artifacts on fetal signal recordings.

Main Methods:

• A joint multiple subspace-based blind source separation (BSS) approach utilizing tensor decomposition across multiple datasets.
• Estimation of the coefficient matrix by incorporating coupled information from stacked covariance matrices of multiple datasets.
• Combining extracted components across four different datasets to enhance signal separation.

Main Results:

• The proposed BSS method demonstrated high accuracy in extracting fetal heart rate (HR) for individual datasets.
• Numerical results showed superior performance compared to conventional signal processing methods in artifact reduction.
• The approach effectively leverages cross-dataset dependencies for improved fetal signal isolation.

Conclusions:

• The developed tensor decomposition-based BSS approach offers a robust solution for non-invasive fetal heart rate monitoring.
• This method significantly improves the accuracy of fetal disease detection by mitigating common signal interferences.
• The joint multiple subspace approach shows promise for enhanced fetal health assessment through cleaner fECG signals.