Simplified Grid Implementation of Medical Image Processing Algorithms using a Workflow Managment System
Dagmar Krefting, Michal Vossberg, Thomas Tolxdorff · 2008
Complex analysis of large amounts of medical image data quickly exceeds storage capacity and computing power of single workstations or small local networks. When limited hardware resources impede full utilization of medical image analysis, a possible solution is the usage of computing grids, the collaboration of distributed resources across institutional borders. Many existing image processing problems would benefit from parallel processing, e.g. on single image or volume slice level. Such coarse-grained parallelization can easily be achieved by implementation into a grid infrastructure. Furthermore, grids allow distributed users to share their code, promoting collaborative projects. In this paper, we describe the grid implementation of existing code using grid workflows. The workflow management system is able to execute all tasks related to grid communication, such as authorization, scheduling and monitoring. It remains to the developer to make the code accessible for the workflow manager, and to define, where the code is found on the grid and what to do with it. We describe the procedure how to bring the code to the grid and show exemplarily the implementation of segmentation and registration algorithms for transrectal ultrasound guided prostate biopsies.