Search

Predictive models and simulations of cardiac function require accurate representations of anatomy, often to the scale of local myocardial fiber structure. However, acquiring this information in a patient-specific manner is challenging. Moreover, the impact of physiological variability in fiber orientation on simulations of cardiac activation is poorly understood. To explore these effects, we implemented bi-ventricular activation simulations using rule-based fiber algorithms and robust uncertainty quantification techniques to generate detailed maps of model variability. Specifically, we utilized polynomial chaos expansion, enabling efficient exploration with reduced computational demand through an emulator function approximating the underlying forward model. Our study focused on examining the epicardial activation sequences of the heart in response to six stimuli locations and five metrics of activation. Our findings revealed that physiological variability in fiber orientation does not significantly affect the location of activation features, but it does impact the overall spread of activation. We observed low variability near the earliest activation sites, but high variability across the rest of the epicardial surface. We conclude that the level of accuracy of myocardial fiber orientation required for simulation depends on the specific goals of the model and the related research or clinical goals.

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.

Recent improvements in detecting acute myocardial ischemia via noninvasive body surface recordings have been driven by modern machine learning. While extensive research has been done using single and 12 lead ECGs, almost no models have incorporated body surface potential mappings. We created two contrasting machine learning models, logistic regression and XGBoost Classifier, and trained them on experimentally acquired body surface mappings with ground truth ischemia measurements recorded from within the heart. These models achieved a mean accuracy of 96.46% and 97.63%, as well as a mean AUC of 0.9927 and 0.9972 for the Logistic Regression and XGBoost classifiers, respectively. The anatomical location and relative contribution of each electrode were visualized and ranked. Then, new models were trained using data from only the top 12, 8, and 3 electrodes. These models trained on only a subset of the electrodes still exhibited relatively high accuracy and AUC, although at much faster training times .

Fiber structure governs the spread of excitation in the heart; however, little is known about the effects of physiological variability in fiber orientation on epicardial activation. To investigate these effects, we implemented ventricular simulations of activation using rule-based fiber orientations, and robust uncertainty quantification algorithms to capture detailed maps of model sensitivity. Specifically, we implemented polynomial chaos expansion, which allows for robust exploration with reduced computational demand through an emulator function to approximate the underlying forward model. We applied these techniques to examine the activation sequence of the heart in response to both epicardial and endocardial stimuli within the left ventricular free wall and variations in fiber orientation. Our results showed that physiological variation in fiber orientation does not significantly impact the location of activation features, but it does impact the overall spread of activation. Future studies will investigate under which circumstances physiological changes in fiber orientation might alter electrical propagation such that the resulting simulations produce misleading outcomes.

Computational models of myocardial ischemia are parameterized using assumptions of tissue properties and physiological values such as conductivity ratios in cardiac tissue and conductivity changes between healthy and ischemic tissues. Understanding the effect of uncertainty in these parameter selections would provide useful insight into the performance and variability of the modeling outputs. Recently developed uncertainty quantification tools allow for the application of polynomial chaos expansion uncertainty quantification to such bioelectric models in order to parsimoniously examine model response to input uncertainty. We applied uncertainty quantification to examine reconstructed extracellular potentials from the cardiac passive bidomain based on variation in the conductivity values for the ischemic tissue. We investigated the model response in both a synthetic dataset with simulated ischemic regions and a dataset with ischemic regions derived from experimental recordings. We found that extracellular longitudinal and intracellular longitudinal conductivities predominately affected simulation output, with the highest standard deviations in regions of extracellular potential elevations. We found that transverse conductivity had almost no effect on model output.

Fiber structure governs the spread of excitation in the heart, however, little is known about the effects of physiological variability in the fiber orientation on epicardial activation. To investigate these effects, we used computer simulation to compare ventricular activation sequences initiated from stimulus sites at regularly spaced depths within the myocardium under varying rule-based fiber ranges. We compared the effects using four characteristics of epicardial breakthrough (BKT): location, area, shape (calculated via the axis ratio of a fitted ellipse), and orientation. Our results showed changes in the BKT characteristics as pacing depth increased, e.g., the area increased, the shape became more circular, and the orientation rotated counterclockwise, regardless of the fiber orientation. Furthermore, the maximal differences in epicardial activation from a single pacing site for location, area, axis ratio, and orientation were 1.2 mm, 74 mm 2 , 0.16, and 26°, respectively. Our results suggest that variability in fiber orientation has a negligible effect on the location, area, and shape of the BKT, while fluctuations were observed in the BKT orientation in response to the fiber fields, especially for epicardial stimulation sites. Our results suggest the fiber field orientation plays only a minor role in activation simulations of ectopic beats.

Acute myocardial ischemia compromises the ordered electrical activation of the heart, however, because of sampling limitations, volumetric changes in activation have not been measured. We used a large-animal experimental model and high-resolution volumetric mapping to study the effects of ischemia on conduction speeds (CS) throughout the myocardium. We estimated CS and electrocardiographic changes (ST segments) and evaluated the spatial and temporal correlations between them across 11 controlled episodes. We found that ischemia induces significant conduction slowing, reducing the global median speed by 25 cm/s. Furthermore, there was a high temporal correlation between the development of ischemic severity and CS (corr. = 0.93) through each episode. The spatial correlations between ST-segment changes and CS slowing were more spatially complex than expected with substantial slowing at the periphery of the zones that showed ST-segment changes. This is the first study that has documented in an experimental model volumetric changes of CS during acute myocardial ischemia and explored the relationships between ischemia development in space and time. We showed that conduction speed changes are spatiotemporally correlated to ischemic severity and illustrated the biphasic response long proposed from cellular studies.

Electrocardiographic imaging (ECGI) systems are still plagued by a myriad of controllable and uncontrollable sources of error, which makes studying and improving these systems difficult. To mitigate these errors, we developed a novel experimental preparation using a rigid pericardiac cage suspended in a torso-shaped electrolytic tank. The 256-electrode cage was designed to record signals 0.5-1.0 cm above the entire epicardial surface of an isolated heart. The cage and heart were fixed in a 192-electrode torso tank filled with electrolyte with predetermined conductivity. The resulting signals served as ground truth for ECGI performed using the boundary element method (BEM) and method of fundamental solutions (MFS) with three regularization techniques: Tikhonov zero-order (Tik0), Tikhonov second-order (Tik2), truncated singular value decomposition (TSVD). Each ECGI regularization technique reconstructed cage potentials from recorded torso potentials well with spatial correlation above 0.7, temporal correlation above 0.8, and root mean squared error values below 0.7 mV. The earliest site of activation was best identified by MFS using Tik0, which localized it to within a range of 1.9 and 4.8 cm. Our novel experimental preparation has shown unprecedented agreement with simulations and represents a new standard for ECGI validation studies.

Introduction: Electrocardiographic imaging (ECGI) requires a model of the torso, and inaccuracy in the position of the heart is a known source of error. We previously presented a method to localize the heart when body and heart surface potentials are known. The goal of this study is to extend this approach to only use body surface potentials., Methods: We used an iterative coordinate descent optimization to estimate the positions of the heart for several consecutive heartbeats relying on the assumption that the epicardial potential sequence is the same in each beat. The method was tested with data synthesized using measurements from a isolated-heart, torso-tank preparation. Improvement was evaluated in terms of both heart localization and ECGI accuracy., Results: The geometric correction resulted in cardiac geometries closely matching ground truth geometry. ECGI accuracy increased dramatically by all metrics using the corrected geometry., Discussion: Future studies will employ more realistic animal models and then human subjects. Success could impact clinical ECGI by reducing errors from respiratory movement and perhaps decrease imaging requirements, reducing both cost and logistical difficulty of ECGI, widening clinical applicability.

Tricuspid regurgitation (TR) is a failure in right-sided AV valve function which, if left untreated, leads to marked cardiac shape changes and heart failure. However, the specific right ventricular shape changes resulting from TR are unknown. The goal of this study is to characterize the RV shape changes of patients with severe TR. RVs were segmented from CINE MRI images. Using particle-based shape modeling (PSM), a dense set of homologous landmarks were placed with geometric consistency on the endocardial surface of each RV, via an entropy-based optimization of the information content of the shape model. Principal component analysis (PCA) identified the significant modes of shape variation across the population. These modes were used to create a patient-prediction model. 32 patients and 6 healthy controls were studied. The mean RV shape of TR patients demonstrated increased sphericity relative to controls, with the three most dominant modes of variation showing significant widening of the short axis of the heart, narrowing of the base at the RV outflow tract (RVOT), and blunting of the RV apex. By PCA, shape changes based on the first three modes of variation correctly identified patient vs. control hearts 86.5% of the time. The shape variation may further illuminate the mechanics of TR-induced RV failure and recovery, providing potential targets for therapies including novel devices and surgical interventions.

Electrocardiographic imaging (ECGI) based detection of myocardial ischemia requires an accurate formulation of the source model, which includes a relationship between extracellular and transmembrane potentials (TMPs). In this study, we used high-resolution intramural experimental recordings and forward modeling to examine the relationship between extracellular potentials and TMPs during myocardial ischemia. We measured extracellular electro-grams from intramural plunge needle arrays during seven controlled ischemia episodes in an animal model. We used three TMP source representations: (1) parameterized and distance-based (defined previously), (2) extracellular-based linear transform, and (3) extracellular-based sigmoidal transform. TMPs for each source formulation were then used to compute extracellular potentials by calculating the passive bidomain forward solution throughout the myocardium. We compared measured and computed potentials. Linear and sigmoidal approaches produced improved results compared to the parameterized method. The RMSE, SC, and TC of linear, sigmoidal, and parameterized methods were 0.85 mV, 1.21 mV, and 3.37 mV; 0.91, 0.88, and 0.47; 0.90, 0.77, and 0.33 respectively. We found extracellular-based calculations of TMPs produced superior forward computations compared to parameterized zones.

Cardiac simulations have become increasingly accurate at representing physiological processes. However, simulations often fail to capture the impact of parameter uncertainty in predictions. Uncertainty quantification (UQ) is a set of techniques that captures variability in simulation output based on model assumptions. Although many UQ methods exist, practical implementation can be challenging. We created UncertainSCI, a UQ framework that uses polynomial chaos (PC) expansion to model the forward stochastic error in simulations parameterized with random variables. UncertainSCI uses non-intrusive methods that parsimoniously explores parameter space. The result is an efficient, stable, and accurate PC emulator that can be analyzed to compute output statistics. We created a Python API to run UncertainSCI, minimizing user inputs needed to guide the UQ process. We have implemented UncertainSCI to: (1) quantify the sensitivity of computed torso potentials using the boundary element method to uncertainty in the heart position, and (2) quantify the sensitivity of computed torso potentials using the finite element method to uncertainty in the conductivities of biological tissues. With UncertainSCI, it is possible to evaluate the robustness of simulations to parameter uncertainty and establish realistic expectations on the accuracy of the model results and the clinical guidance they can provide.

Introduction: Changes in conduction velocity are indicative of a wide variety of cardiac abnormalities yet measuring conduction velocity is challenging, especially within the myocardial volume. In this study we investigated a novel technique to reconstruct activation fronts and estimate three-dimensional (3D) conduction velocity (CV) from experimental intramural recordings., Methods: From the intermittently sampled electrograms we both reconstruct the activation profile and compute the reciprocal of the gradient of activation times and a series of streamlines that allows for the CV estimation., Results: The reconstructed activation times agreed closely with simulated values, with 50% to 70% of the nodes ≤ 1ms of absolute error. We found close agreement between the CVs calculated using reconstructed versus simulated activation times. Across the reconstructed stimulation sites we saw that the reconstructed CV was on average 3.8% different than the ground truth CV., Discussion: This study used simulated datasets to validate our methods for reconstructing 3D activation fronts and estimating conduction velocities. Our results indicate that our method allows accurate reconstructions from sparse measurements, thus allowing us to examine changes in activation induced by experimental interventions such as acute ischemia, ectopic pacing, or drugs.