Fetal QRS extraction from single-channel abdominal ECG using adaptive improved permutation entropy

• Published in: Australasian physical & engineering sciences in medicine Vol. 47; no. 2; pp. 563 - 573
• Main Authors: Mansourian, Nastaran, Sarafan, Sadaf, Torkamani-Azar, Farah, Ghirmai, Tadesse, Cao, Hung
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.06.2024; Springer Nature B.V
• Subjects: Abdomen - diagnostic imaging; Adaptive algorithms; Algorithms; Annotations; Biological and Medical Physics; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Biophysics; Computational efficiency; Electrocardiography; Entropy; Female; Fetal Monitoring; Fetus - diagnostic imaging; Humans; Medical and Radiation Physics; Monitoring; Permutations; Pregnancy; Scientific Paper; Signal Processing, Computer-Assisted; Wearable computers; Wearable technology; Signal processing; Fetal QRS extraction; Adaptive improved permutation entropy (AIPE); Fetal monitoring
• ISSN: 2662-4729; 0158-9938; 2662-4737; 2662-4737; 1879-5447
• DOI: 10.1007/s13246-024-01386-0

Fetal electrocardiogram (fECG) monitoring is crucial for assessing fetal condition during pregnancy. However, current fECG extraction algorithms are not suitable for wearable devices due to their high computational cost and multi-channel signal requirement. The paper introduces a novel and efficient algorithm called Adaptive Improved Permutation Entropy (AIPE), which can extract fetal QRS from a single-channel abdominal ECG (aECG). The proposed algorithm is robust and computationally efficient, making it a reliable and effective solution for wearable devices. To evaluate the performance of the proposed algorithm, we utilized our clinical data obtained from a pilot study with 10 subjects, each recording lasting 20 min. Additionally, data from the PhysioNet 2013 Challenge bank with labeled QRS complex annotations were simulated. The proposed methodology demonstrates an average positive predictive value ( + P ) of 91.0227%, sensitivity ( Se ) of 90.4726%, and F 1 score of 90.6525% from the PhysioNet 2013 Challenge bank, outperforming other methods. The results suggest that AIPE could enable continuous home-based monitoring of unborn babies, even when mothers are not engaging in any hard physical activities.