Skip to content
Microsystems, Electronics and Acoustics
ISSN 2523-4447 · e‑ISSN 2523-4455 Open Access · CC BY-SA 4.0
Vol. 31 · Issue 2 · 2026 Aug 4, 2026 Electronic Systems and Signals

A A Complex Method for Non-invasive Fetal Electrocardiogram Signal Extraction and Cardiac Arrhythmia Classification

DS
Dmytro A. Shevchuk Corresponding National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Ukraine
NI
PhD, Assoc. Prof. Nataliia Heorhiivna Ivanushkina National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Ukraine
Pages356487.1-356487.11 PublishedAug 4, 2026 LicenseOpen Access
pdf (Ukrainian)
VOL 31 · ISS 2 · 2026 View issue

Abstract

This study is devoted to the development of a complex method for extracting the fetal electrocardiogram (fECG) from non-invasive abdominal recordings and for the automated classification of fetal cardiac arrhythmias using machine learning techniques. The early diagnosis of fetal arrhythmias is challenged by a low signal-to-noise ratio in abdominal signals due to the dominance of maternal cardiac activity and the presence of strong physiological interference. To solve this problem, a combined approach integrating multichannel statistical signal separation and spectral analysis of heart rate variability is proposed. The method consists of two stages. At the first stage, fECG extraction is performed using frequency filtering and blind source separation techniques, including principal component analysis (PCA) and independent component analysis (ICA), combined with iterative suppression of the maternal electrocardiogram based on singular value decomposition (SVD). At the second stage, sequences of RR intervals are derived and transformed into a hybrid feature vector combining time-domain heart rate variability metrics and spectral Burg autoregressive coefficients. Numerical results showed that an autoregressive model order of p = 6 provided the best classification performance without indications of overfitting. For the task of binary classification of fetal states (normal or arrhythmia), a unified classification model was developed based on a Support Vector Machine (SVM) classifier combined with a patient-level voting strategy. The proposed method was validated using publicly available datasets: the Abdominal and Direct Fetal ECG Database (ADFECGDB) for assessing signal extraction quality and the Non-Invasive Fetal ECG Arrhythmia Database (NIFEADB) for training and testing the classifier. Evaluated using a Leave-One-Patient-Out (LOPO) cross-validation scheme, the proposed approach achieved an F1-score of 83,3%. Future work will be focused on expanding the diagnostic capabilities of the system by analyzing a broader spectrum of fetal arrhythmias, transitioning to multiclass classification, and applying synthetically generated data to overcome the shortage of annotated fetal pathological records.

Keywords

#abdominal electrocardiogram #fetal electrocardiogram #arrhythmia #blind source separation #autoregressive processes #machine learning

References

  1. Z. Alfirevic, G. M. Gyte, A. Cuthbert, and D. Devane, “Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour,” Cochrane Database of Systematic Reviews, vol. 2019, no. 5, Feb. 2017, DOI: https://doi.org/10.1002/14651858.CD006066.pub3 .
  2. T. Kupka et al., “New Method for Beat-to-Beat Fetal Heart Rate Measurement Using Doppler Ultrasound Signal,” Sensors, vol. 20, no. 15, p. 4079, Jul. 2020, DOI: https://doi.org/10.3390/s20154079 .
  3. G. Aggarwal and Y. Wei, “Non-Invasive Fetal Electrocardiogram Monitoring Techniques: Potential and Future Research Opportunities in Smart Textiles,” Signals, vol. 2, no. 3, pp. 392–412, Jun. 2021, DOI: https://doi.org/10.3390/signals2030025 .
  4. E. W. Abdulhay, R. J. Oweis, A. M. Alhaddad, F. N. Sublaban, M. A. Radwan, and H. M. Almasaeed, “Non-Invasive Fetal Heart Rate Monitoring Techniques,” Biomedical Science and Engineering, vol. 2, no. 3, pp. 53–67, Aug. 2014, DOI: https://doi.org/10.12691/BSE-2-3-2 .
  5. Z. Li and X. Guo, “Assessment of Fetal Well-Being,” in Electronic Fetal Monitoring, Singapore: Springer Singapore, 2021, pp. 1–18. DOI: https://doi.org/10.1007/978-981-15-7364-4 .
  6. W. Xueyun and Z. Wei, “Application of kernel PCA for foetal ECG estimation,” Electron Lett, vol. 54, no. 6, pp. 340–342, Mar. 2018, DOI: https://doi.org/10.1049/el.2018.0071 .
  7. R. Martinek et al., “Comparative Effectiveness of ICA and PCA in Extraction of Fetal ECG From Abdominal Signals: Toward Non-invasive Fetal Monitoring,” Front Physiol, vol. 9, May 2018, DOI: https://doi.org/10.3389/fphys.2018.00648 .
  8. Y. S. Alshebly and M. Nafea, “Isolation of Fetal ECG Signals from Abdominal ECG Using Wavelet Analysis,” IRBM, vol. 41, no. 5, pp. 252–260, Oct. 2020, DOI: https://doi.org/10.1016/j.irbm.2019.12.002 .
  9. K. Barnova et al., “A novel algorithm based on ensemble empirical mode decomposition for non-invasive fetal ECG extraction,” PLoS One, vol. 16, no. 8, p. e0256154, Aug. 2021, DOI: https://doi.org/10.1371/journal.pone.0256154 .
  10. R. Kahankova et al., “A Review of Signal Processing Techniques for Non-Invasive Fetal Electrocardiography,” IEEE Rev Biomed Eng, vol. 13, pp. 51–73, 2020, DOI: https://doi.org/10.1109/RBME.2019.2938061 .
  11. R. Jaros, R. Martinek, and R. Kahankova, “Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal,” Sensors, vol. 18, no. 11, p. 3648, Oct. 2018, DOI: https://doi.org/10.3390/s18113648 .
  12. S. Aziz, S. Ahmed, and M.-S. Alouini, “ECG-based machine-learning algorithms for heartbeat classification,” Sci Rep, vol. 11, no. 1, p. 18738, Sep. 2021, DOI: https://doi.org/10.1038/s41598-021-97118-5
  13. J. Jezewski, A. Matonia, T. Kupka, D. Roj, and R. Czabanski, “Abdominal and Direct Fetal ECG Database (ADFECGDB),” PhysioNet, version 1.0.0, 2012. [Online]. DOI: https://doi.org/10.13026/C2RP4B
  14. J. A. Behar, L. Bonnemains, V. Shulgin, J. Oster, O. Ostras, and I. Lakhno, “Non-Invasive Fetal ECG Arrhythmia Database (NIFEA DB),” PhysioNet, version 1.0.0. [Online]. DOI: https://doi.org/10.13026/C2CT0S
  15. R. J. Martis, U. R. Acharya, and L. C. Min, “ECG beat classification using PCA, LDA, ICA and Discrete Wavelet Transform,” Biomed Signal Process Control, vol. 8, no. 5, pp. 437–448, Sep. 2013, DOI: https://doi.org/10.1016/j.bspc.2013.01.005 .
  16. V. Ionescu, “Fetal ECG Extraction from Multichannel Abdominal ECG Recordings for Health Monitoring During Labor,” Procedia Technology, vol. 22, pp. 682–689, 2016, DOI: https://doi.org/10.1016/j.protcy.2016.01.143 .
  17. Y. Li, X. Luo, H. Zhao, J. Cui, Y. She, D. Li, L. Jiang, and X. Zhang, “Noninvasive Extraction of Maternal and Fetal Electrocardiograms Using Progres-sive Periodic Source Peel-off,” arXiv preprint arXiv: 2406.01281, 2024, DOI: https://doi.org/10.48550/arXiv.2406.01281 .
  18. A. Veduta et al., “Treatment of Fetal Arrhythmias,” J Clin Med, vol. 10, no. 11, p. 2510, Jun. 2021, DOI: https://doi.org/10.3390/jcm10112510 .
  19. J. A. Behar, L. Bonnemains, V. Shulgin, J. Oster, O. Ostras, and I. Lakhno, “Noninvasive fetal electrocardiography for the detection of fetal arrhythmias,” Prenat Diagn, vol. 39, no. 3, pp. 178–187, Feb. 2019, DOI: https://doi.org/10.1002/pd.5412 .
  20. A. Ouelli, B. E. B.Elhadadi, H. A. H.Aissaoui, and B. B. B.Bouikhalene, “AR Modeling for Cardiac Arrhythmia Classification using MLP Neural Networks,” Int J Comput Appl, vol. 47, no. 24, pp. 44–51, Jul. 2012, DOI: https://doi.org/10.5120/7508-0638 .
  21. T. Nield, “Essential Math for Data Science: Take Control of Your Data with Fundamental Linear Algebra, Probability, and Statistics” O’Reilly Media, Inc., 2022, ISBN: 978-1-098-10293-7.
  22. J. Hao, Y. Yang, Z. Zhou, and S. Wu, “Fetal Electrocardiogram Signal Extraction Based on Fast Independent Component Analysis and Singular Value Decomposition,” Sensors, vol. 22, no. 10, p. 3705, May 2022, DOI: https://doi.org/10.3390/s22103705 .
  23. Md. S. R. Pavel, Md. R. Islam, and A. M. Siddiqee, “Fetal Arrhythmia Detection Using Fetal ECG Signal,” in 2019 IEEE International Conference on Telecommunications and Photonics (ICTP), IEEE, Dec. 2019, pp. 1–4. DOI: https://doi.rg/10.1109/ICTP48844.2019.9041789 .
  24. X. F. Joseph, A. T. Waktola, D. Senay, and S. R. Breesha, “An Enhanced Classification Technique for Fetal ECG Signal Separation to Diagnose Fetal Heart Diseases,” in 2019 IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing (INCOS), IEEE, Apr. 2019, pp. 1–5. DOI: https://doi.rg/10.1109/INCOS45849.2019.8951317 .
  25. X. Shi, K. Yamamoto, T. Ohtsuki, Y. Matsui, and K. Owada, “Unsupervised Learning-Based Non-Invasive Fetal ECG Muti-Level Signal Quality Assessment,” Bioengineering, vol. 10, no. 1, p. 66, Jan. 2023, DOI: https://doi.rg/10.3390/bioengineering10010066 .
  26. G. A. Pradipta, P. Desiana Wulaning Ayu, M. Liandana, and D. P. Hostiadi, “Enhanced Fetal Arrhythmia Classification by Non-Invasive ECG Using Cross Domain Feature and Spatial Differences Windows Information,” IEEE Access, vol. 13, pp. 6729–6749, 2025, DOI: https://doi.rg/10.1109/ACCESS.2025.3526586 .

License

CCBY-SA 4.0
Creative Commons Attribution 4.0 International

This work is openly licensed — share and adapt freely with attribution to the authors and the journal. View license terms ↗

Copyright (c) 2026 Дмитро А. Шевчук, канд. техн. наук, доц. Наталія Георгіївна Іванушкіна (Автор)

§ 06 — Related

Similar articles in this journal

Related peer-reviewed studies published in this journal.
View all issues

Quasistatic Identification of The Shading Structure of Solar Panels by The Method of Cardinal Number Vectors

PhD, Assoc. Prof. Olena Olehivna Abakumova, Dr.Sc.(Eng.), Prof. Valerii Yakovych Zhuikov, Mykola O. Lukianov

In the article the problem of identification the shading structure of solar panels is considered. The relevance of the study is conditioned by the necessity of rapid and…

#solar panels #photovoltaic surface #shadow structure modeling #quasistatic identification
p. 362283.1-362283.7
10.20535/2523-4455.mea.362283

Simulation Modeling of a Filter-Compensating Converter for a Three-Phase Three-Wire Network

Ivan Andriiovych Vakulenko, PhD, Assoc. Prof. Dmytro Anatoliiovych Mykolaiets, PhD, Assoc. Prof. Kateryna Serhiivna Klen

This paper addresses the critical issue of power quality degradation in modern power systems caused by the proliferation of non-linear loads. The study focuses on the…

#filter-compensating converter #three-phase three-wire network #power quality #total harmonic distortion
p. 361138.1-361138.6
10.20535/2523-4455.mea.361138

Modeling the Impact of Temperature on the Electrical Activity of Cardiac Cells: Identification of Key Ionic Currents

Vitalii O. Ivashchuk, PhD, Assoc. Prof. Nataliia Heorhiivna Ivanushkina

This study presents the development of a modified electrophysiological model of a human epicardial cardiomyocyte, achieved by taking into account the temperature effects…

#temperature effects #cardiac electrophysiology #sensitivity analysis #Morris method
p. 357181.1-357181.13
10.20535/2523-4455.mea.357181

Most read articles by the same author(s)

Similar Articles

1-10 of 35

You may also start an advanced similarity search for this article.