Computing in cardiology最新文献

Identification of patient-specific scar and fibrosis is a critical step in the personalization of cardiac computational models. Late gadolinium enhanced cardiac magnetic resonance imaging (LGE-cMRI) is often used to identify patient anatomy, as well as tissue fibrosis and scar. Automated methods to identify scar from LGE-cMRI exist. Still, there is no clear consensus as to which is best in the context of patient-specific computational modeling of atrial fibrillation. There has been no substantial investigation into the effects that variability in scar may have on downstream patient-specific simulations. This study compares the distribution of scar patterns generated via automated LGE-cMRI analysis alongside human-guided scar identification. We assess the effects each identified scar pattern has on downstream computational modeling outputs by comparing the number of stable re-entrant arrhythmias induced In Silico in atrial fibrillation. We find both substantial disagreement between scar patterns identified via automated and human-guided methods, as well as sensitivity in the arrhythmia simulation outcomes across scar patterns. These results highlight the sensitivity of such computational models to these input parameters and enforce the need for robust personalization tools in the cardiac modeling field.

Ventricular arrhythmia frequently complicates myocardial ischemic events, sometimes to devastating ends. Accurate arrhythmia prediction in this setting could improve outcomes, yet traditional models struggle with the temporal complexity of the data. This study employs a Long Short-Term Memory (LSTM) network to predict the time to the next premature ventricular contraction (PVC) using high-resolution experimental data. We analyzed electrograms from 11 large animal experiments, identifying 1832 PVCs, and computed time-to-PVC. An LSTM model (247 inputs, 1024 hidden units) was trained on 10 experiments, with one held out for testing, achieving a validation MAE of 8.6 seconds and a test MAE of 135 seconds (loss 68.5). Scatter plots showed strong validation correlation and a positive test trend, suggesting the potential of this approach.

Acute myocardial ischemia (AMI) is one of the leading causes of cardiovascular deaths around the globe. Yet, clinical early detection and patient risk stratification of AMI remain an unmet need, in part due to poor performance of traditional electrocardiogram (ECG) interpretation. Machine learning (ML) techniques have shown promise in analysis of ECGs, even detecting cardiac diseases not identifiable via traditional analysis. However, there has been limited usage of ML tools in the case of AMI due to a lack of high-quality training data, especially detailed ECG recordings throughout the evolution of ischemic events. In this study, we applied ML to predict the ischemic tissue volume directly from body surface ECGs in an AMI animal model. The developed ML networks performed favorably, with an average R² value of 0.932 suggesting a robust prediction. The study also provides insights on how to create and utilize ML tools to enhance clinical risk stratification of patients experiencing AMI.

Simulating the electrical behavior of the heart requires accounting for parameter errors, model inaccuracies, and individual variations in settings, which can all be influenced by user choices or disease conditions. To map the effects of parameter uncertainty, we built on previous findings employing bi-ventricular activation simulations and robust uncertainty quantification (UQ) techniques based on polynomial chaos expansion (PCE) that maps variability in propagation simulations. The PCE approach offers efficient stochastic exploration with reduced computational demands. To ensure reliable results, we focused here on the importance of testing for polynomial order and sample size, aiming to obtain accurate outcomes with minimal computational burden. Order testing involves determining the polynomial degree used for calculating statistics, whereas sample testing pertains to identifying the necessary number and values of the parameters from which the UQ model is estimated. The guide for both steps was to ensure consistency in the results, roughly emulating a convergence analysis. We applied this approach to a bi-ventricular activation simulation using UncertainSCI and quantified the effects of physiological variability in conduction velocity. Our results show that the selection of the appropriate polynomial degree order and sample dataset influences the outcomes of simulations and should be a required step before performing a UQ analysis.

The mechanisms that drive and sustain atrial fibrillation (AF) continue to be a highly researched topic. Atrial fibrosis has been linked with increased incidence of AF and conduction, but how fibrosis may lead to AF sustaining remains unknown. Our study aims to highlight heterogeneity in atrial fibrosis and how differences in fibrotic architecture may influence the sustainability of AF. In our study, we utilize a transgenic goat model with cardiac-specific over-expression of TGFβ-1 gene to examine structural differences of the fibrotic regions between animals that are inducible for AF and animals that remain AF-free. Our results indicate that there are structural differences between the fibrotic regions of AF inducible and non-inducible animals. Animals inducible for AF were found to have increased structural isotropy and increased fiber disarray within the fibrotic regions. Histology samples taken from the fibrotic regions showed fibrotic strands disrupted the tissue fibers in a more obstructive manner in the inducible animal group. These results highlight the heterogeneous differences of fibrotic regions.

Because serum potassium and ECG morphology changes exhibit a well-understood connection, and the timeline of ECG changes can be relatively quick, there is motivation to explore the sensitivity of ML based prediction of serum potassium using 12 lead ECG data with respect to the time between the ECG and potassium readings. We trained a convolutional neural network to classify abnormal (serum potassium above 5 mEq/L) vs normal (serum potassium between 4 and 5 mEq/L) from the ECG alone. We compared training with ECGs and potassium measurements filtered to be within 1 hour, 30 minutes, and 15 minutes of each other. We explored scenarios that both leveraged all available data at each time cutoff as well as restricted data to match training set sizes across the time cutoffs. For each case, we trained five separate instances of our neural network to account for variability. The 1 hour cutoff with all data resulted in an average area under the receiver operator curve (AUC) of 0.850 and a weighted accuracy of 76.3%, 15 minutes resulted in 0.814, 72.5%, and 30 minutes. Truncating the training sets to the same size as the 15 minute cutoff results in comparable average accuracy and AUC for all. Our future studies will continue to explore the performance of ML potassium predictions through investigations of failure cases, identification of biases, and explainability analyses.

The electrocardiogram (ECG) is the most common clinical tool to measure the electrical activity of the heart. Despite its ubiquity and utility, traditional ECG analysis methods are limited to primarily human interpretation. Machine learning tools can be employed to automate detection of diseases, and to detect patterns that are not available to traditional ECG analysis. However, contemporary machine learning tools are limited by requirements for large labeled datasets, which can be scarce for rare diseases. Self-supervised learning (SSL) can address this data scarcity. We implemented the momentum contrast (MoCo) framework, a form of SSL, using a large clinical ECG dataset. We then assessed the learning using Low Left Ventricular Ejection Fraction (LVEF) detection as the downstream task. We compared the SSL improvement of LVEF classification across different input augmentations. We observed that optimal augmentation hyperparameters varied substantially based on the training dataset size, indicating that augmentation strategies may need to be tuned based on problem and dataset size.

Cardiac computational models are powerful tools to improve treatment of complex cardiac arrhythmias. However, such computational models rely on many uncertain inputs, and the effects of this input uncertainty on the model-derived treatment strategies are unclear. We have developed a computational model-guided ablation planning tool to aid in the ablation of reentrant circuits found in atypical left atrial flutter (ALAF). We then applied parametric uncertainty quantification to assess the effect of errors and variability in the conductivity of fibrotic tissue on the model outputs and suggested ablation patterns. In a computational model of a patient who presented with ALAF, we found that our model-guided ablation tool reduced the number of simulated ALAF circuits from 10 preablation to 4 postablation. Uncertainty quantification revealed that fibrotic conductivity affected the suggested ablation sites substantially; however, the uncertainty quantification also provided a method to display a proposed ablation strategy in a manner that accounts for the input parameter uncertainty. The results of this study show the twofold insight of UQ. This method provides a robust means to explore the effects of input parameter variability on predictions of reentrant arrhythmia. We suggest it can also present modeling results that display the uncertainty associated with model predictions.

The 12-lead electrocardiogram (ECG) is the most common front-line diagnosis tool for assessing cardiovascular health, yet traditional ECG analysis cannot detect many diseases. Machine learning (ML) techniques have emerged as a powerful set of techniques for producing automated and robust ECG analysis tools that can often predict diseases and conditions not detectable by traditional ECG analysis. Many contemporary ECG-ML studies have focused on utilizing the full 12-lead ECG; however, with the increased availability of single-lead ECG data from wearable devices, there is a clear motivation to explore the development of single-lead ECG-ML techniques. In this study we developed and applied a deep learning architecture for the detection of low left ventricular ejection fraction (LVEF), and compared the performance of this architecture when it was trained with individual leads of the 12-lead ECG to the performance when trained using the entire 12-lead ECG. We observed that single-lead-trained networks performed similarly to the full 12-lead-trained network. We also noted patterns of agreement and disagreement between network low LVEF predictions across the different lead-trained networks.

Individual variability in parameter settings, due to either user selection or disease states, can impact accuracy when simulating the electrical behavior of the heart. Here, we aim to test the impact of inevitable uncertainty in conduction velocities (CVs) on the output of simulations of cardiac propagation, given three stimulus locations on the left ventricular (LV) free wall. To understand the role of physiological variability in CV in simulations of cardiac activation, we generated detailed maps of the variability in propagation simulations by implementing bi-ventricular activation simulations and quantified the effects by deploying robust uncertainty quantification techniques based on polynomial chaos expansion (PCE). PCE allows efficient stochastic exploration with reduced computational demand by utilizing an emulator for the underlying forward model. Our results suggest that CV within healthy physiological ranges plays a small role in the activation times across all stimulation locations. However, we noticed differences in variation coefficients depending on the stimulation site, i.e., LV endocardium, midmyocardium, and epicardium. We observed low levels of variation in activation times near the earliest activation sites, whereas there was higher variation toward the termination sites. These results suggest that CV variability can play a role when simulating healthy and diseased states.