Efficient computation of electrograms and ECGs in human whole heart simulations using a reaction-eikonal model

Area of Science:

• Computational biology
• Cardiac electrophysiology
• Biomedical modeling

Background:

• Bidomain models offer mechanistic insight into cardiac electrical activity but are computationally expensive.
• Personalizing these models for clinical applications is hindered by significant computational demands.
• Accurate replication of patient-specific electrograms and potential maps is crucial for clinical translation.

Purpose of the Study:

• To develop a computationally efficient reaction-eikonal (R-E) model for simulating cardiac extracellular potentials and electrograms.
• To assess the fidelity of the R-E model compared to traditional bidomain models.
• To enable faster, personalized cardiac electrophysiology modeling for potential clinical use.

Main Methods:

• Development of a novel reaction-eikonal (R-E) model.
• Utilized a biventricular human electrophysiology model with a realistic His-Purkinje system.
• Compared R-E model outputs against a high-resolution reaction-diffusion (R-D) bidomain model.

Main Results:

• The R-E model achieved computational savings greater than three orders of magnitude compared to the R-D bidomain model.
• R-E model accurately predicted extracellular potential fields, electrograms, and body surface ECGs.
• High fidelity was maintained despite significant computational efficiency gains.

Conclusions:

• The computationally efficient R-E model accurately simulates cardiac electrical phenomena.
• R-E models are suitable for rapid forward simulations in personalized cardiac modeling.
• This approach facilitates the clinical application of patient-specific electrophysiology models.