Related Experiment Video
Updated: Jul 4, 2026
04:36
Noninvasive Electrocardiography in the Perinatal Mouse
Published on: June 12, 2020
Comparison study of population-based methods for non-invasive fetal electrocardiography extraction
Akshaya Raj1, Jindrich Brablik1, Radana Vilimkova Kahankova1
1Department of Cybernetics and Biomedical Engineering, Faculty of Electrical Engineering and Computer Science, VSB-Technical University of Ostrava, Ostrava, Czechia.
Frontiers in Medicine
|July 3, 2026
Summary
This study compared five algorithms for extracting non-invasive fetal electrocardiogram (NI-fECG) signals. Gray wolf optimization, moth flame optimization, particle swarm optimization, and whale optimization algorithm showed similar, stable performance, outperforming the artificial bee colony algorithm.
Area of Science:
- Biomedical Signal Processing
- Computational Intelligence
- Maternal-Fetal Medicine
Background:
- Extracting non-invasive fetal electrocardiogram (NI-fECG) from abdominal signals is challenging due to maternal ECG (mECG) interference.
- Overlapping time and frequency domains of fECG and mECG necessitate advanced signal processing techniques.
Purpose of the Study:
- To comparatively analyze the performance of five population-based optimization algorithms for NI-fECG extraction.
- To evaluate the efficacy of artificial bee colony (ABC), gray wolf optimization (GWO), moth flame optimization (MFO), particle swarm optimization (PSO), and whale optimization algorithm (WOA) in fECG signal isolation.
Main Methods:
- Five population-based algorithms (ABC, GWO, MFO, PSO, WOA) were integrated with sequential analysis (SA) for fECG extraction.
- The algorithms were tested on Labor and Pregnancy datasets, with performance evaluated by R-peak detection accuracy.
- Experiments were repeated 30 times independently to assess algorithm stability.
Main Results:
- GWO, MFO, PSO, and WOA demonstrated comparable and stable performance in NI-fECG extraction.
- The ABC algorithm exhibited poor performance and instability compared to the other tested algorithms.
- R-peak detection accuracy served as the primary metric for evaluating extraction efficiency.
Conclusions:
- GWO, MFO, PSO, and WOA are promising for NI-fECG extraction, offering robust and accurate results.
- The ABC algorithm requires further modifications or hybrid approaches to improve its effectiveness and stability for fECG signal processing.
- The study highlights the potential of population-based algorithms in improving non-invasive fetal monitoring techniques.
