Toward GPGPU accelerated human electromechanical cardiac simulations.
Bottom Line: Specifically, we port to the GPU a number of components of CHeart--a CPU-based finite element code developed for simulating multi-physics problems.Speedup of up to 72 × compared with SC and 2.6 × compared with MC was also observed for the PDE solve.Using the same human geometry, the GPU implementation of mechanics residual/Jacobian computation provided speedups of up to 44 × compared with SC and 2.0 × compared with MC.
Affiliation: Department of Biomedical Engineering, King's College London, UK.Show MeSH
Related in: MedlinePlus
Mentions: For mechanics problem, we have simulated the model described in Section 4.2, and we have used the same LV human geometry as for the electrophysiology problem. However, in this case, it has been discretized on the basis of a coarser mesh (see Figure 2(c)), reflecting the type of meshes often observed in cardiac mechanics (although the results illustrated are expected to be consistent with larger cardiac mechanics meshes). To solve the mechanics problem on this mesh, we have mapped the activation time from the fine grid electrophysiology mesh onto the mechanics mesh. Using these activation time values, we simulated the cardiac cycle, comparing performance over a single beat (with a duration of 1 s and a time step of 0.001 ms). Figure 5 shows displacement values during a cycle simulation. Figure 6 shows the principal strain vectors and fibers at end diastole and mid systole steps.
Affiliation: Department of Biomedical Engineering, King's College London, UK.