GPU-based iterative tomography

Abstract: 

Even though iterative reconstruction algorithms give very accurate results, their long computation time inhibits practical applications. The most computationally intensive steps of the algorithms used by the ASTRA research group are very well parallelizable. Modern GPU cores consist of many processing units and are optimized for operations also required during tomographic reconstructions. With the release of NVIDIA CUDA, GPUs became fully programmable. Our research is focused on maximizing parallelism in previously developed algorithms and developing new hardware platforms to allow for practical applications of iterative reconstruction algorithms.

Publications: 
S. van der Maar, K. J. Batenburg, and J. Sijbers, "Experiences with Cell-BE and GPU for tomography", Embedded COmputer Systems: Architectures, Modeling, and Simulation - 9th International Workshop, SAMOS 2009 - Proceedings: Springer-Verlag Berlin Heidelberg, pp. 298-307, July, 2009.
Main researcher(s): 
Research Area: 
Research keywords: