Noninvasive Fetal Electrocardiography: Models, Technologies, and Algorithms

Sameni, Reza

doi:10.1007/978-3-030-54403-4_5

Cited by 6 publications

(4 citation statements)

References 76 publications

(152 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The impact of factors (A) and (B) above, are less explicit. However, looking at the entries (2, 2) and (3,3) of the Fisher Information and CRLB matrices of the SoG model in ( 26)-( 29), we notice that the CRLB (Fisher Information) is proportional (inversely proportional) to the effective width of the guassian model. Therefore, the rule of thumb is that the estimation error variance of ECG wave centers and widths is smaller (larger) for narrower (wider) waves.…”

Section: Discussionmentioning

confidence: 99%

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Fattahi¹,

Sameni²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

<div>Objective: Clinical parameter estimation from the electrocardiogram (ECG) is a recurrent field of research. It is debated that ECG parameter estimation performed by human experts and machines/algorithms is always model-based (implicitly or explicitly). Therefore, depending on the selected data-model, the adopted estimation scheme (least-squares error, maximum likelihood, or Bayesian), and the prior assumptions on the model parameters and noise distributions, any estimation algorithm used in this context has an upper performance bound, which is not exceedable (for the same model and assumptions).</div><div><br></div><div>Method: In this research, we develop a comprehensive theoretical framework for ECG parameter estimation and derive the Cramér-Rao lower bounds (CRLBs) for the most popular signal models used in the ECG modeling literature; namely bases expansions (including polynomials) and sum of Gaussian functions.</div><div><br></div><div>Results: The developed framework is evaluated over real and synthetic data, for three popular applications: T/R ratio estimation, ST-segment analysis and QT-interval estimation, using the state-of-the-art estimators in each context, and compared with the derived theoretical CRLBs.</div><div>Conclusion and Significance: The proposed framework and the derived CRLBs provide fact-based guidelines for the selection of data-models, sampling frequency (beyond the Nyquist requirements), modeling segment length, the number of beats required for average ECG beat extraction, and other factors that influence the accuracy of ECG-based clinical parameter estimation.</div>

show abstract

Section: Discussionmentioning

confidence: 99%

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Fattahi¹,

Sameni²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For decades, a significant branch of biomedical research has been devoted to the reliable extraction of clinical parameters from the ECG. In recent years, technologies such as wearable healthcare gadgets [1], ECG acquisition during stress-tests and sports [2], and noninvasive fetal ECG monitoring have emerged [3], which have once again brought about the necessity of accurate ECG parameter extraction, especially in edge-computing and highly noisy scenarios.…”

Section: Introductionmentioning

confidence: 99%

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Fattahi

Sameni

2022

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

Clinical parameter estimation from the electrocardiogram (ECG) is a recurrent field of research. It is debated that ECG parameter estimation by human experts and machines/algorithms is always model-based (implicitly or explicitly). Therefore, all estimation algorithms used in this context have performance bounds in terms of the achievable mean squared error, which are not exceedable. These bounds depend on the adopted data-model, the estimation scheme (least-squares error, maximum likelihood, or Bayesian), and prior assumptions on the model parameters and noise distributions.In this research, we develop a comprehensive theoretical framework for ECG parameter estimation and derive the Cramér-Rao lower bounds (CRLBs) for the most popular signal models used in the ECG modeling literature, namely functional expansions (including polynomials) and sum of Gaussian functions. The developed framework is evaluated over real and synthetic data for three popular applications: T-to-R wave ratio estimation, ST-segment analysis and QT-interval estimation, using the state-of-the-art estimators in each context. The proposed framework and the derived CRLBs provide practical guidelines for the selection of data-models, sampling frequency (beyond the Nyquist rate), modeling segment length, number of beats required for ECG beat averaging, and other factors that influence the accuracy of ECG-based clinical parameter estimation.

show abstract

“…Besides, when the number of signal channels is less than the number of signal sources, it is called a low-rank. There are also a variety of processing algorithms for low-rank channels aECG signals (Sameni 2021). Typical examples are the extraction of fetal ECG from time-varying and low-rank non-invasive abdominal records of pregnant women (Fahimeh et al 2018) and the separation of fetal and maternal components of low-dimensional maternal aECG signal (Mantas and Vaidotas 2020) et al.…”

Section: Introductionmentioning

confidence: 99%

Method of Extracting Fetal QRS for Single-Lead Abdominal ECG Signal Based On Combined Compressive Sensing and Clustering Analysis

Dong

et al. 2021

Preprint

View full text Add to dashboard Cite

Objective: The fetal electrocardiogram (ECG) is an objective index that reflects a fetus’s health status. Non-invasive abdominal ECG (aECG) was obtained by placing silicone electrodes on pregnant women’s abdominal wall. However, fetal QRS (fQRS) extraction is very challenging due to maternal ECG interference, motion artifacts, and other noise. Approach: This paper introduces a new single-lead non-invasive fQRS extraction method based on compressive sensing and clustering analysis. This method can be applied to portable, low-power remote fetal ECG (fECG) acquisition equipment based on the Internet of Things (IoT). It is mainly divided into the following steps: (1) optimal component extraction of single-channel signal based on compressive sensing theory; (2) location of maternal QRS (mQRS) using the clustering method based on extreme value; (3) maternal ECG (mECG) elimination; (4) The preliminary location of fQRS based on double clustering and the correction of fQRS based on fetal RR interval. Main results: The new algorithm proposed in this paper is verified on two publicly available data sets. The averages of these indicators are Se=98.53%, PPV=98.28%, ACC=96.95%, F1=98.43% for the Silesia datasets and Se=97.59%, PPV=97.63%, ACC=95.44%, F1=97.62% for the Challenge datasets A. Significance: The results show that it is feasible and reliable to locate fQRS from a single-channel aECG signal under the condition of reducing power consumption. It lays a foundation for implementing the low-power wireless transmission of fECG signal and remote fetal heart rate (FHR) monitoring based on the IoT.

show abstract

Noninvasive Fetal Electrocardiography: Models, Technologies, and Algorithms

Cited by 6 publications

References 76 publications

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Cramér-Rao Lower Bounds of Model-Based Electrocardiogram Parameter Estimation

Method of Extracting Fetal QRS for Single-Lead Abdominal ECG Signal Based On Combined Compressive Sensing and Clustering Analysis

Contact Info

Product

Resources

About