Combining endocardial mapping and electrocardiographic imaging (ECGI) for improving PVC localization: A feasibility...

Area of Science:

• Biomedical Engineering
• Cardiac Electrophysiology
• Medical Imaging

Background:

• Accurate cardiac activation wavefront reconstruction is vital for diagnosing and treating arrhythmias.
• Intramural myocardial activation mapping has been a significant challenge due to limitations in traditional electrical mapping techniques.
• Recent advances in electrocardiographic imaging (ECGI) enable both endocardial and epicardial activation mapping.

Purpose of the Study:

• To develop and validate a novel technique for reconstructing intramural activation times using a combined epicardial and endocardial approach.
• To assess the accuracy and robustness of this combined mapping technique for intramural activation reconstruction.

Main Methods:

• Utilized 11 simulations of ventricular activation with pacing from various myocardial wall sites.
• Extracted endocardial and epicardial activation maps at clinically relevant resolutions.
• Employed thin-plate-spline radial basis functions for interpolating activation times through the myocardium.
• Evaluated reconstruction accuracy using root-mean-squared error (RMSE) and percentage of nodes within 1 ms of ground truth.

Main Results:

• The combined approach achieved a root-mean-squared error (RMSE) of 3 ms and correctly identified 70% of nodes within 1 ms of ground truth.
• In the most challenging scenarios, the RMSE was 4 ms, with 60% of nodes accurately reconstructed within 1 ms.
• Demonstrated robust intramural activation reconstruction across diverse stimulation sites.

Conclusions:

• A combined endocardial and epicardial mapping strategy effectively reconstructs intramural cardiac activation wavefronts.
• This integrated ECGI approach offers a significant advancement for understanding and treating cardiac arrhythmias.
• The method provides accurate and robust intramural activation time reconstructions, overcoming previous limitations.