EP Workbench: A computational platform for identifying fibrotic regions and conduction disturbances in the atria using conduction velocity.

Background: Fibrotic remodelling in the atria, associated with atrial fibrillation (AF), creates areas of slowed and heterogenous conduction. These areas act as substrates that promote AF maintenance and perpetuation. Hence, detecting these regions is crucial for understanding and managing AF. One promising approach for identifying these regions is the estimation of conduction velocity (CV).

Purpose: In this study, we sought to enhance OpenEP Workbench software for researchers by (1)incorporating three well established method for CV estimation (triangulation, planar fitting, and radial basis function interpolation) and CV divergence calculation to detect conduction disturbances; (2)assess the performance of these three CV calculation methods in identifying fibrotic regions, using simulated data as the ground truth (3)developing a visualisation tool to enable identification of slow conduction regions at the optimal classification threshold attained from (2).

Results: The classification performance of the three CV methods in identifying fibrotic regions were assessed, resulting with the highest accuracy and AUC for the triangulation method (accuracy=78%, AUC=0.88). EP Workbench was enhanced with tools to create 3D surface maps of the generated voltage, CV and divergence parametrised voltage using electro-anatomical mapping data. Additionally, a histogram analysis tool was implemented to identify regions of slow conduction velocity from electroanatomic mapping data. Using a threshold of <0.3m/s, two slow conducting regions can be visualised on the posterior wall of a test case.

Conclusion: The enhanced OpenEP Workbench provides a pre-built pipeline for researcher and clinicians to optimise CV and fibrotic region classifiers and quantify conduction velocity heterogeneity in electro-anatomical mapping data.