DG-Mapping: a novel software package for the analysis of any type of reentry and focal activation of simulated, experimental or clinical data of cardiac arrhythmia

Improved algorithm for generating evenly-spaced streamlines from an orientation field on a triangulated surface

BACKGROUND

Vector fields such as cardiac fiber orientation can be visualized on a surface using streamlines. The application of evenly-spaced streamline generation to the construction of interconnected cable structure for cardiac propagation models has more stringent requirements imperfectly fulfilled by current algorithms.

METHOD

We developed an open-source C++/python package for the placement of evenly-spaced streamlines on a triangulated surface. The new algorithm improves upon previous works by more accurately handling streamline extremities, U-turns and limit cycles, by providing stronger geometrical guarantees on inter-streamline minimal distance, particularly when a high density of streamlines (up to 10μm spacing) is desired, and by making a more efficient parallel implementation available. The approach requires finding intersections between geometrical capsules and triangles to update an occupancy mask defined on the triangles. This enables fast streamline integration from thousands of seed points to identify optimal streamline placement.

RESULTS

The algorithm was assessed qualitatively on different left atrial models of fiber orientation with varying mesh resolutions (up to 375k triangles) and quantitatively by measuring streamline lengths and distribution of inter-streamline minimal distance. The complexity and the computational performance of the algorithm were studied as a function of streamline spacing in relation to triangular mesh resolution.

CONCLUSION

More accurate geometrical computations, attention to details and fine-tuning led to an algorithm more amenable to applications that require precise positioning of streamlines.

A Review of Personalised Cardiac Computational Modelling Using Electroanatomical Mapping Data

Computational models of cardiac electrophysiology have gradually matured during the past few decades and are now being personalised to provide patient-specific therapy guidance for improving suboptimal treatment outcomes. The predictive features of these personalised electrophysiology models hold the promise of providing optimal treatment planning, which is currently limited in the clinic owing to reliance on a population-based or average patient approach. The generation of a personalised electrophysiology model entails a sequence of steps for which a range of activation mapping, calibration methods and therapy simulation pipelines have been suggested. However, the optimal methods that can potentially constitute a clinically relevant in silico treatment are still being investigated and face limitations, such as uncertainty of electroanatomical data recordings, generation and calibration of models within clinical timelines and requirements to validate or benchmark the recovered tissue parameters. This paper is aimed at reporting techniques on the personalisation of cardiac computational models, with a focus on calibrating cardiac tissue conductivity based on electroanatomical mapping data.

Directed graph mapping shows rotors maintain non-terminating and focal sources maintain self-terminating Torsade de Pointes in canine model

Torsade de Pointes is a polymorphic ventricular tachycardia which is as yet incompletely understood. While the onset of a TdP episode is generally accepted to be caused by triggered activity, the mechanisms for the perpetuation is still under debate. In this study, we analysed data from 54 TdP episodes divided over 5 dogs (4 female, 1 male) with chronic atrioventricular block. Previous research on this dataset showed both reentry and triggered activity to perpetuate the arrhythmia. 13 of those TdP episodes showed reentry as part of the driving mechanism of perpetuating the episode. The remaining 41 episodes were purely ectopic. Reentry was the main mechanism in long-lasting episodes (>14 beats), while focal sources were responsible for maintaining shorter episodes. Building on these results, we re-analysed the data using directed graph mapping This program uses principles from network theory and a combination of positional data and local activation times to identify reentry loops and focal sources within the data. The results of this study are twofold. First, concerning reentry loops, we found that on average non-terminating (NT) episodes (≥10 s) show significantly more simultaneous reentry loops than self-terminating (ST) TdP (<10 s). Non-terminating episodes have on average 2.72 ± 1.48 simultaneous loops, compared to an average of 1.33 ± 0.66 for self-terminating episodes. In addition, each NT episode showed a presence of (bi-)ventricular loops between 10.10% and 69.62% of their total reentry duration. Compared to the ST episodes, only 1 in 4 episodes (25%) showed (bi-)ventricular reentry, lasting only 7.12% of its total reentry duration. This suggests that while focal beats trigger TdP, macro-reentry and multiple simultaneous localized reentries are the major drivers of long-lasting episodes. Second, using heatmaps, we found focal sources to occur in preferred locations, instead of being distributed randomly. This may have implications on treatment if such focal origins can be disabled reliably.

Directed Graph Mapping for Ventricular Tachycardia: A Comparison to Established Mapping Techniques

BACKGROUND

Understanding underlying mechanism(s) and identifying critical circuit components are fundamental to successful ventricular tachycardia (VT) ablation. Directed graph mapping (DGM) offers a novel technique to identify the mechanism and critical components of a VT circuit.

OBJECTIVES

This study sought to evaluate the accuracy of DGM in VT ablation compared with traditional mapping techniques and a commercially available automated conduction velocity mapping (ACVM) tool.

METHODS

Patients with structural heart disease who had undergone a VT ablation with entrainment-proven critical isthmus and a high-density electroanatomical activation map were included. Traditional mapping (TM) consisted of a combination of local activation time and entrainment mapping and was considered the gold standard for determining the VT mechanism, circuit, and isthmus location. The same local activation time values were then processed using DGM and a commercially available ACVM (Coherent Mapping, Biosense Webster) tool. The aim of this study was to compare TM vs DGM and ACVM in their ability to identify the VT mechanism, characterize the VT circuit, and locate the critical isthmus.

RESULTS

Thirty-five cases were identified. TM classified the VT mechanism as focal in 7 patients and re-entrant in 28 patients. TM classified 11 VTs as single-loop re-entry, 15 as dual-loop re-entry, 1 as complex, and 1 case was indeterminant. The overall agreement between DGM and TM for determining VT mechanism and circuit type was strong (kappa value = 0.79; P < 0.01), as was the agreement between ACVM and TM (kappa value = 0.66; P < 0.01). Both DGM and ACVM identified the putative VT isthmus in 25 (89%) of the re-entrant cases. Focal activation was correctly identified by both techniques in all cases.

CONCLUSIONS

DGM is a rapid automated algorithm that has a strong level of agreement with TM for manually re-annotated VT maps.