Problem
Clinical use of aneurysm-treatment simulation requires confidence that predicted flow changes match both experiments and established solvers. I evaluated lattice Boltzmann variants for unsteady intracranial aneurysm hemodynamics.
Approach
- Simulate pulsatile flow in complex patient-derived intracranial aneurysm geometries.
- Compare multiple LBM implementations against PIV/LDA measurements.
- Benchmark against a commercial finite-volume solver.
- Assess grid convergence and runtime characteristics, including GPU acceleration.
Key finding
A calibrated LBM configuration reproduced experimental and finite-volume trends with competitive accuracy while reducing runtime substantially with GPU execution.
Why it matters
Reliable and faster virtual-treatment flow prediction can make patient-specific planning more feasible in real clinical timelines.
Outputs
- Publication details are listed in the References section below.
- Comparative CFD and validation visuals are provided on this page.