GPU Parallel Hemodynamics via Lattice Boltzmann
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current patient-specific computational hemodynamics methods are limited by computational expense, inefficiency due to lack of parallelization, and inaccuracies introduced by data reconstruction and mesh generation when using macroscopic Navier-Stokes solvers and external image processing software.
Innovation Solution
A method utilizing a GPU parallel-computation framework that employs mesoscale models, specifically the simplified lattice Boltzmann method (SLBM), volumetric lattice Boltzmann method (VLBM), and lattice spring method (LSM) for non-invasive quantification of in vivo blood flow and flow-artery interaction, integrating image data processing, fluid dynamics, and structural mechanics on a unified platform.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If macroscopic Navier-Stokes solvers are used to solve fluid dynamics, then fluid dynamics can be solved, but computational expense increases and parallel acceleration efficiency is reduced
Solution Approach 1:
The patent replaces macroscopic Navier-Stokes solvers with a mesoscale lattice Boltzmann method (LBM). The LBM uses kinetic theory-based particle distribution functions instead of continuous macroscopic equations, enabling efficient parallel computation on GPUs while maintaining accuracy in modeling blood flow and fluid-structure interaction in complex arterial geometries.
Solution Approach 2:
The patent changes the computational scale from macroscopic to mesoscale by introducing a unified mesoscale framework. This parameter change allows the system to model both fluid dynamics and structural mechanics at the same scale, improving computational efficiency through unified mesoscale equations that can be parallelized effectively.
2Measurement precision
If external image processing software is used to extract anatomical structures, then anatomical structures can be extracted, but computation efficiency is limited due to lack of parallelization
Solution Approach 1:
The patent merges image processing, anatomical structure extraction, and computational simulation into a single unified mesoscale platform. This integration eliminates the need for separate external software tools and enables parallel processing of the entire workflow, from CT/MRI image input to hemodynamic simulation output, significantly improving computational efficiency.
Solution Approach 2:
The unified mesoscale platform performs multiple functions including image processing, anatomical segmentation, mesh generation, and fluid-structure interaction simulation within a single system. This multi-functional approach eliminates data transfer between separate software tools and enables end-to-end parallel computation.
3Ease of manufacture
If data reconstruction and mesh generation are performed separately, then the gap between image processing and CFD solver can be filled, but extra time is consumed and conversion errors and inaccuracy are introduced
Solution Approach 1:
The patent combines data reconstruction and mesh generation with the fluid-structure interaction simulation in a unified mesoscale framework. Anatomical structures are directly extracted from images and used to define computational domains without separate mesh generation steps, eliminating conversion errors and maintaining data accuracy throughout the simulation process.
Data Source
AI summary
A method for computing patient-specific hemodynamics. The method includes receiving three dimensional imaging data of a patent, extracting anatomical data from the three dimensional imaging data, calculating velocity and pressure fields corresponding to the extracted anatomical data, and calculating displacement and velocity of extracted solid particles corresponding to the anatomical data. The anatomical data comprises an anatomical boundary.