Understanding the Utility of Endocardial Electrocardiographic Imaging in Epi-Endocardial Mapping of 3D Reentrant Circuits.

Background: Studies of VT mechanisms are largely based on a 2D portrait of reentrant circuits on one surface of the heart. This oversimplifies the 3D circuit that involves the depth of the myocardium. Simultaneous epicardial and endocardial (epi-endo) mapping was shown to facilitate a 3D delineation of VT circuits, which is however difficult via invasive mapping. Objective: This study investigates the capability of noninvasive epicardial-endocardial electrocardiographic imaging (ECGI) to elucidate the 3D construct of VT circuits, emphasizing the differentiation of epicardial, endocardial, and intramural circuits and to determine the proximity of mid-wall exits to the epicardial or endocardial surfaces. Methods: 120-lead ECGs of VT in combination with subject-specific heart-torso geometry are used to compute unipolar electrograms (CEGM) on ventricular epicardium and endocardia. Activation isochrones are constructed, and the percentage of activation within VT cycle length is calculated on each surface. This classifies VT circuits into 2D (surface only), uniform transmural, nonuniform transmural, and mid-myocardial (focal on surfaces). Furthermore, the endocardial breakthrough time was accurately measured using Laplacian eigenmaps, and by correlating the delay time of the epi-endo breakthroughs, the relative distance of a mid-wall exit to the epicardium or the endocardium surfaces was identified. Results: We analyzed 23 simulated and in-vivo VT circuits on post-infarction porcine hearts. In simulated circuits, ECGI classified 21% as 2D and 78% as 3D: 82.6% of these were correctly classified. The relative timing between epicardial and endocardial breakthroughs was correctly captured across all cases. In in-vivo circuits, ECGI classified 25% as 2D and 75% as 3D: in all cases, circuit exits and entrances were consistent with potential critical isthmus delineated from combined LGE-MRI and catheter mapping data. Conclusions: ECGI epi-endo mapping has the potential for fast delineation of 3D VT circuits, which may augment detailed catheter mapping for VT ablation.

Hybrid Neural State-Space Modeling for Supervised and Unsupervised Electrocardiographic Imaging.

State-space modeling (SSM) provides a general framework for many image reconstruction tasks. Error in a priori physiological knowledge of the imaging physics, can bring incorrectness to solutions. Modern deep-learning approaches show great promise but lack interpretability and rely on large amounts of labeled data. In this paper, we present a novel hybrid SSM framework for electrocardiographic imaging (ECGI) to leverage the advantage of state-space formulations in data-driven learning. We first leverage the physics-based forward operator to supervise the learning. We then introduce neural modeling of the transition function and the associated Bayesian filtering strategy. We applied the hybrid SSM framework to reconstruct electrical activity on the heart surface from body-surface potentials. In unsupervised settings of both in-silico and in-vivo data without cardiac electrical activity as the ground truth to supervise the learning, we demonstrated improved ECGI performances of the hybrid SSM framework trained from a small number of ECG observations in comparison to the fixed SSM. We further demonstrated that, when in-silico simulation data becomes available, mixed supervised and unsupervised training of the hybrid SSM achieved a further 40.6% and 45.6% improvements, respectively, in comparison to traditional ECGI baselines and supervised data-driven ECGI baselines for localizing the origin of ventricular activations in real data.

Improving Generalization by Learning Geometry-Dependent and Physics-Based Reconstruction of Image Sequences.

Deep neural networks have shown promise in image reconstruction tasks, although often on the premise of large amounts of training data. In this paper, we present a new approach to exploit the geometry and physics underlying electrocardiographic imaging (ECGI) to learn efficiently with a relatively small dataset. We first introduce a non-Euclidean encoding-decoding network that allows us to describe the unknown and measurement variables over their respective geometrical domains. We then explicitly model the geometry-dependent physics in between the two domains via a bipartite graph over their graphical embeddings. We applied the resulting network to reconstruct electrical activity on the heart surface from body-surface potentials. In a series of generalization tasks with increasing difficulty, we demonstrated the improved ability of the network to generalize across geometrical changes underlying the data using less than 10% of training data and fewer variations of training geometry in comparison to its Euclidean alternatives. In both simulation and real-data experiments, we further demonstrated its ability to be quickly fine-tuned to new geometry using a modest amount of data.

Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data.

OBJECTIVE: This work investigates the possibility of disentangled representation learning of inter-subject anatomical variations within electrocardiographic (ECG) data. METHODS: Since ground truth anatomical factors are generally not known in clinical ECG for assessing the disentangling ability of the models, the presented work first proposes the SimECG data set, a 12-lead ECG data set procedurally generated with a controlled set of anatomical generative factors. Second, to perform such disentanglement, the presented method evaluates and compares deep generative models with latent density modeled by nonparametric Indian Buffet Process to account for the complex generative process of ECG data. RESULTS: In the simulated data, the experiments demonstrate, for the first time, concrete evidence of the possibility to disentangle key generative anatomical factors within ECG data in separation from task-relevant generative factors. We achieve a disentanglement score of 92.1% while disentangling five anatomical generative factors and the task-relevant generative factor. In both simulated and real-data experiments, this work further provides quantitative evidence for the benefit of disentanglement learning on the downstream clinical task of localizing the origin of ventricular activation. Overall, the presented method achieves an improvement of around 18.5%, and 11.3% for the simulated dataset, and around 7.2%, and 3.6% for the real dataset, over baseline CNN, and standard generative model, respectively. CONCLUSION: These results demonstrate the importance as well as the feasibility of the disentangled representation learning of inter-subject anatomical variations within ECG data. SIGNIFICANCE: This work suggests the important research direction to deal with the well-known challenge posed by the presence of significant inter-subject variations during an automated analysis of ECG data.