Extraction for fetal ECG using single channel blind source separation algorithm based on multi-algorithm fusion

He, Peng; Chen, Xiao Meng; Liang, Yan; Zeng, Hongbin

doi:10.1051/matecconf/20164401026

Cited by 8 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It uses the bootstrap method, the Hough method, the Akaike information criterion method and principal component analysis (PCA) to estimate the independent component number. Then, the four numbers obtained are fused utilizing particle swarm optimization (PSO) to determine the accurate number [ 70 ].…”

Section: Single Channel Signal Sourcesmentioning

confidence: 99%

See 1 more Smart Citation

Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal

Jaroš

Martínek

Kahánková

2018

Sensors

View full text Add to dashboard Cite

Fetal electrocardiography is among the most promising methods of modern electronic fetal monitoring. However, before they can be fully deployed in the clinical practice as a gold standard, the challenges associated with the signal quality must be solved. During the last two decades, a great amount of articles dealing with improving the quality of the fetal electrocardiogram signal acquired from the abdominal recordings have been introduced. This article aims to present an extensive literature survey of different non-adaptive signal processing methods applied for fetal electrocardiogram extraction and enhancement. It is limiting that a different non-adaptive method works well for each type of signal, but independent component analysis, principal component analysis and wavelet transforms are the most commonly published methods of signal processing and have good accuracy and speed of algorithms.

show abstract

Section: Single Channel Signal Sourcesmentioning

confidence: 99%

“…He et al [ 70 ] developed the SCBSS method. For the evaluation, they used a single channel signal mixed with four man-made signals and the MIT-BIH arrhythmia database [ 67 ], where two ECG signals were chosen.…”

Section: Single Channel Signal Sourcesmentioning

confidence: 99%

Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal

Jaroš

Martínek

Kahánková

2018

Sensors

View full text Add to dashboard Cite

show abstract

“…The focus on this paper is to demonstrate the feasibility of source separation in ECG signals in order to detect R peaks, using a blind source separation algorithm called Non-negative Matrix Factorization (NMF). Some researches have been made combining electrocardiogram and NMF, such as ECG-EMG separation [5] or extraction of fetal ECG [6], but to the best of our knowledge, none on the separation of the ECG waves.…”

Section: Introductionmentioning

confidence: 99%

R-Peak Detection in Holter ECG Signals Using Non-Negative Matrix Factorization

Guyot¹,

Voiriot²,

Djermoune³

et al. 2018

2018 Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

Holter monitoring is mainly used for medical followup and diagnosis of patients with suspected cardiac arrhythmia such as heart rhythm irregularities that can be missed during classical electrocardiogram recording (ECG). However, these long-term continuous recordings represent a large amount of data that cannot be processed by hand. In this article, we present a new method based on Non-negative Matrix Factorization (NMF) to detect Rpeaks in Holter signals. The approach consists in two stages: source separation based on the different timefrequency patterns of the QRS complexes and the other waves of the signal (P and T waves) and R-peak detection using Automatic Objective Thresholding (AOT). The proposed approach is validated on the MIT-BIH Arrhythmia database and achieves an average sensitivity of 99.59% and a precision of 99.69%. Using the MIT-BIH Noise Stress Test database, we also show the ability of our approach to discriminate R-peaks in signals contaminated with different noises.

show abstract

“…Based on the Blind Source Separation (BSS) technique, He et al proposed a method of mECG and fECG separation from a single channel abdominal signal. They also used the Empirical Mode Decomposition (EMD) method to decompose the collected single channel signal into multiple Intrinsic Mode Function (IMF) mapping single channels [18]. Ghazdali et al proposed an approach for denoising of the observed ECG signals using a Bilateral Total Variation (BTV) filter and then minimizing the Kullback-Leibler divergence [19] between copula densities to separate the fECG signal from the mECG [20].…”

Section: Introductionmentioning

confidence: 99%

Quantification of Feto-Maternal Heart Rate from Abdominal ECG Signal Using Empirical Mode Decomposition for Heart Rate Variability Analysis

2017

View full text Add to dashboard Cite

Abstract:In this paper, a robust method of feto-maternal heart rate extraction from the non-invasive composite abdominal Electrocardiogram (aECG) signal is presented. The proposed method is based on the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) method, in which a composite aECG signal is decomposed into its constituent frequency components called Intrinsic Mode Functions (IMFs) or simply "modes", with better spectral separation. Decomposed IMFs are then selected manually according to probable maternal and fetal heart rate information and are processed further for quantification of maternal and fetal heart rate and variability analysis. The proposed method was applied to aECG recordings collected from three different sources: (i) the PhysioNet (adfecgdb) database; (ii) the PhysioNet (nifecgdb) database; and (iii) synthetic aECG signal generated from mathematical modeling in the LabVIEW software environment. An overall sensitivity of 98.83%, positive diagnostic value of 97.97%, accuracy of 96.93% and performance index of 96.75% were obtained in the case of Maternal Heart Rate (MHR) quantification, and an overall sensitivity of 98.13%, positive diagnostic value of 97.62%, accuracy of 95.91% and performance index of 95.69% were obtained in case of Fetal Heart Rate (FHR) quantification. The obtained results confirm that CEEMDAN is a very robust and accurate method for extraction of feto-maternal heart rate components from aECG signals. We also conclude that non-invasive aECG is an effective and reliable method for long-term FHR and MHR monitoring during pregnancy and labor. The requirement of manual intervention while selecting the probable maternal and fetal components from "n" number of decomposed modes limits the real-time application of the proposed methodology. This is due to the fact that the number of modes "n" produced by the CEEMDAN decomposition is unpredictable. However, the proposed methodology is well suited for applications where a small time-delay or offset in feto-maternal monitoring can be acceptable. In future, application-specific modification of the CEEMDAN algorithm can be implemented to eliminate manual intervention completely and will be suitable for long-term feto-maternal monitoring.

show abstract

Extraction for fetal ECG using single channel blind source separation algorithm based on multi-algorithm fusion

Cited by 8 publications

References 12 publications

Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal

Non-Adaptive Methods for Fetal ECG Signal Processing: A Review and Appraisal

R-Peak Detection in Holter ECG Signals Using Non-Negative Matrix Factorization

Quantification of Feto-Maternal Heart Rate from Abdominal ECG Signal Using Empirical Mode Decomposition for Heart Rate Variability Analysis

Contact Info

Product

Resources

About