Abstract
We present and compare a variety of parallelization approaches for a real-world case study on modern parallel and distributed computer architectures. Our case study is a production-quality, time-intensive algorithm for medical image reconstruction used in computer tomography (PET). We parallelize this algorithm for the main kinds of contemporary parallel architectures: shared-memory multiprocessors, distributed-memory clusters, graphics processing units (GPU) using the CUDA framework, the Cell processor and, finally, how various architectures can be accessed in a distributed Grid environment. The main contribution of the paper, besides the parallelization approaches, is their systematic comparison regarding four important criteria: performance, programming comfort, accessibility, and cost-effectiveness. We report results of experiments on particular parallel machines of different architectures that confirm the findings of our systematic comparison.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
IBM Inc (2010) Software Development Kit for Multicore Acceleration Version 3.0. http://www.ibm.com/developerworks/power/cell/
NVIDIA Corp (2010) NVIDIA CUDA compute unified device architecture. http://developer.nvidia.com/object/cuda.html
Brune C, Sawatzky A, Burger M (2009) Bregman-EM-TV methods with application to optical nanoscopy. In: Proceedings of the 2nd international conference on scale space and variational methods in computer vision. Lectures notes in computer science, vol 5567. Springer, Berlin, pp 235–246
Hoefler T, Schellmann M, Gorlatch S, Lumsdaine A (2008) Communication optimization for medical image reconstruction algorithms. In: Recent advances in parallel virtual machine and message passing interface. Lectures notes in computer science, vol 5205. Springer, Berlin, pp 75–83
Kösters T, Wübbeling F, Natterer F (2006) Scatter correction in PET using the transport equation. In: IEEE nuclear science symposium and medical imaging conference record. IEEE, New York, pp 3305–3309
Meiländer D, Schellmann M, Gorlatch S (2009) Implementing a data-parallel application with low data locality on multicore processors. In: International conference on architecture of computing systems — workshop proceedings. VDE, pp 57–64
Reader AJ, Erlandsson K, Flower MA, Ott RJ (1998) Fast accurate iterative reconstruction for low-statistics positron volume imaging. Phys Med Biol 43(4):823–834
Ryoo S, Rodrigues C, Baghsorkhi S, Stone S, Kirk D, Hwu W (2008) Optimization principles and application performance evaluation of a multithreaded GPU using CUDA. In: PPoPP ’08: proc of the 13th ACM SIGPLAN symposium, pp 73–82
Schäfers KP, Reader AJ, Kriens M, Knoess C, Schober O, Schäfers M (2005) Performance evaluation of the 32-module QuadHIDAC small-animal PET scanner. J Nucl Med 46(6):996–1004
Schellmann M, Böhm D, Wichmann S, Gorlatch S (2007) Towards a grid system for medical image reconstruction. In: IEEE nuclear science symposium and medical imaging conference record. IEEE, New York, pp 3019–3025
Schellmann M, Gorlatch S (2007) Comparison of two decomposition strategies for parallelizing the 3D list-mode OSEM algorithm. In: Proceedings fully 3D meeting and HPIR workshop, pp 37–40
Schellmann M, Vörding J, Gorlatch S, Meiländer D (2008) Cost-effective medical image reconstruction: from clusters to graphics processing units. In: Proceedings of the 2008 conference on computing frontiers. ACM, New York, pp 283–292
Schelmann M, Gorlatch M, Meiländer D, Kösters T, Schäfers K, Wübbeling F, Burger M (2009) Parallel medical image reconstruction: from graphics processors to grids. In: 10th International Conference PaCT-2009. Lectures notes in computer science, vol 5698. Springer, Berlin, pp 457–473
Shepp LA, Vardi Y (1982) Maximum likelihood reconstruction for emission tomography. IEEE Trans Med Imag 1:113–122
Siddon RL (1985) Fast calculation of the exact radiological path for a three-dimensional CT array. Med Phys 12(2):252–255
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Schellmann, M., Gorlatch, S., Meiländer, D. et al. Parallel medical image reconstruction: from graphics processing units (GPU) to Grids. J Supercomput 57, 151–160 (2011). https://doi.org/10.1007/s11227-010-0397-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-010-0397-z