Numerical Solution of Diffusion Models in Biomedical Imaging on Multicore Processors

Abstract

In this paper, we consider nonlinear partial differential equations (PDEs) of diffusion/advection type underlying most problems in image analysis. As case study, we address the segmentation of medical structures. We perform a comparative study of numerical algorithms arising from using the semi-implicit and the fully implicit discretization schemes. Comparison criteria take into account both the accuracy and the efficiency of the algorithms. As measure of accuracy, we consider the Hausdorff distance and the residuals of numerical solvers, while as measure of efficiency we consider convergence history, execution time, speedup, and parallel efficiency. This analysis is carried out in a multicore-based parallel computing environment.

Techniques of medical image processing and analysis accelerated by high-performance computing: a systematic literature review

“…Non-rigid image registration techniques demand lengthy execution times because of the input images are usually large and because the adopted transformation model adopted requires substantially more time to compute and evaluate the similarity measure used. The experiments were performed on an SGI Origin 3800 massively parallel computer, and all the results were compared using different degrees of parallelism (2,16,32, and 48 threads); the performance achieved showed a reduced linear execution time.…”

Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation

“…By using a forward finite differences scheme to discretize the scale derivative, three iterative schemes can be obtained: explicit, semi-implicit and fully-implicit. An analysis of the convergenge and accuracy order of those schemes in our setting is provided in [8]. Here, we use a semi-implicit scheme, since it is invariant to many transformations, such as grey level shift, translation, rotation, etc., and enjoys consistency, convergency and stability properties [4,20].…”