Noninvasive electrocardiographic imaging with low-rank and non-local total variation regularization

Mu, Lide; Liu, Huafeng

doi:10.1016/j.patrec.2020.07.007

Cited by 9 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tikhonov regularisation (Throne andOlson 2000, Kara et al 2019) is one of the most classic L2-norm methods, which only provides a solution with compromise accuracy because of the inherent smoothing constraint in L2-norm. In order to overcome the weakness and synthesize the sparsity and piece-wise smoothness of the potential, the works proposed cardiac electric sparse imaging techniques (Yu et al 2015, Ting et al 2018 and the methods which imposed a sparse-based constraint with L1-norm have been proposed, such as total variation (TV) regularisation (Wang et al 2011, Xu et al 2014, Mu and Liu 2020. Recently, further works which select different forms of constraints or combine more information were later presented to improve the stability and convergence of their solutions.…”

Section: Introductionmentioning

confidence: 99%

“…We realize that some efforts to reconstruct endocardial and epicardial potential (EEP) (Erik et al 2017, Fang et al 2019b from body surface ECGs are related to our work. Notable methods include the use of Twomey regularization (Twomey 1963), low rank and sparse decomposition (LSD) framework (Fang et al 2019a), and nonlocal total variation regularization (Mu and Liu 2020). Recently, some authors have employed a novel dataadaptive regression framework (Onak et al 2021) to obtain a mapping from BSP to EEP without redundant parameters.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Nonlocal based FISTA network for noninvasive cardiac transmembrane potential imaging

Ran,

Cheng,

Xie

et al. 2024

Phys. Med. Biol.

Self Cite

View full text Add to dashboard Cite

Objective: The primary aim of our study is to advance our understanding and diagnosis of cardiac diseases. We focus on the reconstruction of myocardial transmembrane potential (TMP) from body surface potential mapping.Approach: We introduce a novel methodology for the reconstruction of the dynamic distribution of TMP. This is achieved through the integration of convolutional neural networks with conventional optimization algorithms. Specifically, we utilize the subject-specific transfer matrix to describe the dynamic changes in TMP distribution and ECG observations at the body surface. To estimate the TMP distribution, we employ LNFISTA-Net, a learnable non-local regularized iterative shrinkage-thresholding network. The coupled estimation processes are iteratively repeated until convergence.Main results: Our experiments demonstrate the capabilities and benefits of this strategy. The results highlight the effectiveness of our approach in accurately estimating the TMP distribution, thereby providing a reliable method for the diagnosis of cardiac diseases.Significance: Our approach demonstrates promising results, highlighting its potential utility for a range of applications in the medical field. By providing a more accurate and dynamic reconstruction of TMP, our methodology could significantly improve the diagnosis and treatment of cardiac diseases, thereby contributing to advancements in healthcare.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Nonlocal based FISTA network for noninvasive cardiac transmembrane potential imaging

Ran,

Cheng,

Xie

et al. 2024

Phys. Med. Biol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several regularization methods are available (i.e. L1-Norm [13], IRN-MLSQR [14], physiologybased regularization (PBR) [15], Kalman filter-based [16], rank-deficient [17], [18]), of which Tikhonov regularization [19] is the most commonly used to solve this inverse problem. It minimizes the functional…”

Section: Introductionmentioning

confidence: 99%

“…Despite the large number of possible methods to solve the inverse problem of electrocardiography, with more being developed every day [14], [16], [18], [20]- [24], no single algorithm consistently outperforms the rest. Previous studies comparing different methods have found that the optimal algorithm can depend on the data set [25], the level of noise [26], [27], the geometric error present in the data [27] among other factors.…”

Section: Introductionmentioning

confidence: 99%

A Patchwork Method to Improve the Performance of Current Methods for Solving the Inverse Problem of Electrocardiography

Bouhamama

Potse

Bear

et al. 2023

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

Objective: Noninvasive electrocardiographic imaging (ECGI) reconstructs cardiac electrical activity from body surface potential measurements. However, current methods have demonstrated inaccuracies in reconstructing sinus rhythm, and in particular breakthrough sites. This study aims to combine existing inverse algorithms, making the most of their advantages while minimizing their limitations. Method: The "patchwork method" (PM) combines two classical numerical methods for ECGI: the method of fundamental solutions (MFS) and the finite-element method (FEM). We assume that the method with the smallest residual in the predicted torso potentials, computed using the boundary element method (BEM), provides the most accurate solution. The PM selects for each heart node and time step the method whose estimated reconstruction error is smallest. The performance of the PM was evaluated using simulated ectopic and normal ventricular beats. Results: Cardiac potentials and activation maps obtained with the PM (CC=0.63±0.01 and 0.61±0.05 respectively) were more accurate than MFS (CC=0.61±0.01 and 0.48±0.05 respectively), FEM (CC=0.58±0.01 and 0.51±0.02 respectively) or BEM (CC=0.57±0.02 and 0.49±0.02 respectively). The PM also located all epicardial breakthrough sites, whereas the traditional numerical methods usually missed one. Furthermore, the PM showed its robustness and stability in the presence of Gaussian noise added to the torso potentials. Conclusion: The PM overcomes some of the limitations of classical numerical methods, improving the accuracy of mapping important features of activation during sinus rhythm and paced beats. Significance: This novel method for optimizing ECGI solutions opens a new avenue for improving not only ECGI but also other inverse problems.

show abstract

Cardiac Transmembrane Potential Imaging with GCN Based Iterative Soft Threshold Network

Liu

2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Noninvasive electrocardiographic imaging with low-rank and non-local total variation regularization

Cited by 9 publications

References 26 publications

Nonlocal based FISTA network for noninvasive cardiac transmembrane potential imaging

Nonlocal based FISTA network for noninvasive cardiac transmembrane potential imaging

A Patchwork Method to Improve the Performance of Current Methods for Solving the Inverse Problem of Electrocardiography

Cardiac Transmembrane Potential Imaging with GCN Based Iterative Soft Threshold Network

Contact Info

Product

Resources

About