The estimation of conductivity parameters from Electrograms (EGMs) could provide valuable insights into the underlying electrophysiological mechanisms of cardiac arrhythmias. However, given the anisotropic behavior of cardiomyocytes, accurate estimation requires knowledge of the varying myocardial fiber orientation. Despite numerous proposed methods for deriving the fiber direction from EGMs, determining the optimal approach for clinical and research applications remains uncertain. Therefore, this paper aims to identify the most accurate and robust method by evaluating their performance with data from realistic cardiac models.
The mean absolute error of estimation (MAE) of nine fiber direction estimation methods was compared using thirty-six EGMs, and their corresponding local activation time maps. The data was generated from 2D and 3D monodomain cardiac models that simulated the fiber architecture in Bachmann's bundle, the right atrium, and the left ventricle.
The methods based on ellipse-fitting to conduction slowness vectors (ECS) or conduction velocity vectors computed either through polynomial surface fitting (ECVPSF) or by solving a geometrical model of the planar wavefront (ECVGM) demonstrated the highest accuracy. The MAE values of these methods when using data from the 2D model were 5º, 0.79º and 3º, respectively. However, the performance of the methods was susceptible to fiber direction heterogeneity across the myocardial wall due to the intramural signal propagation, resulting in a rotation of the activation wavefront and MAE values up to 37º with data from the 3D models.
To enhance the accuracy of the ECS, ECVPSF, and ECVGM methods, signal transmission within myocardial layers using endocardial recordings should be considered. Additionally, integrating epicardial EGMs recorded after pacing at different sites into these approaches can address the impact of wavefront direction on estimates. These improvements pave the way for developing an accurate and robust fiber direction estimation method, advancing the diagnosis and treatment of cardiac arrhythmias.