Related Experiment Video
Updated: May 4, 2026
09:20
Lumped-Parameter and Finite Element Modeling of Heart Failure with Preserved Ejection Fraction
Published on: February 13, 2021
6.1K
Adaptive step ODE algorithms for the 3D simulation of electric heart activity with graphics processing units.
V M Garcia-Molla1, A Liberos2, A Vidal1
1Department of Information Systems and Computing, Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain.
Computers in Biology and Medicine
|January 1, 2014
Summary
Adaptive and fixed step methods were compared for simulating cardiac electrical activity on GPUs. Adaptive methods offer superior accuracy and robustness, while fixed step methods can be faster for large ordinary differential equation systems.
Area of Science:
- Computational science
- Biomedical engineering
- Numerical analysis
Background:
- Simulating complex biological systems like cardiac electrical activity requires efficient numerical methods.
- Large systems of ordinary differential equations (ODEs) pose computational challenges.
- Graphics Processing Units (GPUs) offer parallel processing capabilities for accelerating simulations.
Purpose of the Study:
- To implement and evaluate adaptive step methods for large ODE systems on GPUs.
- To compare the performance of adaptive versus fixed step methods in this context.
- To assess the accuracy and robustness of different numerical approaches for cardiac electrophysiology simulations.
Main Methods:
- Implementation of adaptive and fixed step ODE solvers on GPUs.
- Application of the Rush-Larsen method to enhance solver efficiency.
- Comparative analysis of simulation speed, accuracy, and robustness.
- Focus on simulating three-dimensional electric cardiac activity.
Main Results:
- Adaptive step methods demonstrated superior accuracy and robustness compared to fixed step methods.
- Fixed step methods were found to be potentially faster in certain scenarios.
- The Rush-Larsen method contributed to improved efficiency across implemented solvers.
Conclusions:
- Adaptive step methods are preferable for accurate and robust simulations of cardiac electrophysiology on GPUs.
- Choice between adaptive and fixed step methods depends on the trade-off between speed, accuracy, and robustness requirements.
- GPU acceleration is effective for large-scale ODE simulations in computational biology.
