A signal quality assessment method for fetal QRS complexes detection

Area of Science:

• Biomedical Engineering
• Signal Processing
• Maternal-Fetal Medicine

Background:

• Non-invasive fetal ECG (NI-FECG) is crucial for fetal health monitoring.
• Signal noise significantly degrades NI-FECG quality, challenging fetal heart rate (FHR) monitoring.
• Existing methods struggle with accurate FHR extraction due to signal interference.

Purpose of the Study:

• To develop and validate a dynamic signal quality assessment algorithm for multi-channel NI-FECG.
• To improve the reliability of FHR monitoring by addressing signal quality issues.
• To enhance the accuracy of fetal QRS (FQRS) complex detection.

Main Methods:

• Proposed a novel algorithm for dynamic signal quality evaluation during FQRS detection.
• Utilized detected FQRS complexes (true R peaks - TRs, false R peaks - FRs) as quality units.
• Estimated channel signal quality based on the accuracy of FQRS detection results.
• Fused TRs from multiple channels for robust FHR monitoring.

Main Results:

• The algorithm effectively selects high-quality signals for FQRS detection.
• Achieved high performance metrics: 97.40% Positive Predictive Value (PPV), 98.33% Sensitivity (SE), and 97.86% F1-score.
• Demonstrated significant improvement in the reliability of multi-channel FHR monitoring.

Conclusions:

• The proposed dynamic signal quality assessment method enhances NI-FECG monitoring.
• This approach offers a promising solution for accurate and reliable FHR monitoring.
• Contributes to advancing the quality assessment techniques in fetal monitoring signals.