Solving the coupled system improves computational efficiency of the bidomain equations.

Area of Science:

• Computational biology
• Biophysics
• Medical imaging

Background:

• The bidomain equations model cardiac action potential propagation.
• Solving these equations at the organ level presents significant computational challenges.
• Current methods often decouple the equations for efficiency, but this may impact accuracy.

Purpose of the Study:

• To compare the efficiency of a coupled bidomain solver with a conventional uncoupled approach.
• To evaluate the impact of solver strategy on computational time and parallel performance.
• To determine the optimal numerical strategy for accurate and efficient cardiac electrophysiology simulations.

Main Methods:

• Implementation and testing of both coupled and uncoupled bidomain equation solvers.
• Utilizing two 3-D rabbit ventricle computational models.
• Performance analysis based on simulation time and parallel scalability.

Main Results:

• The coupled bidomain solver demonstrated up to 80% speedup compared to the uncoupled method.
• The coupled approach exhibited superior parallel performance, especially for larger, more complex models.
• Despite requiring a larger linear system, the coupled method proved more efficient overall.

Conclusions:

• Solving the bidomain equations as a coupled system is more computationally efficient than the conventional uncoupled approach.
• The coupled method offers a viable strategy for accelerating cardiac electrophysiology simulations without compromising accuracy.
• This finding has significant implications for reducing the computational burden of whole-heart modeling.