Advanced Imaging


Advanced Imaging Magazine

Updated: January 12th, 2011 10:01 AM CDT

High Performance 3D Image Reconstruction

Improved performance of high-end platforms accelerates reconstruction times and enhances image quality
3D geometry
3D geometry: One of the first steps consists of re-sampling the raw projections to compensate for source-detector geometry distortions. The projections are up-sampled in such a way that a nearest neighbor-approach can be used during the backprojection. For a given sub-volume, only a small part of the projection needs to be considered and loaded into fast memory for the best performance.
Run-Time Stack chart
The software tools available for the cell platform include all traditional development tools, such as compilers, debuggers and real-time trace analyzers. The run-time software layers make it possible to simplify the multiprocessing paradigm and the required data distribution techniques in such a way that the productivity can be maintained.

By Dr. Marc Kachelriess, Dr. Michael Knaup, and Olivier Bockenbach

By Dr. Marc Kachelriess, Institute of Medical Physics, University of Erlangen-Nuremberg, Germany; Dr. Michael Knaup, RayConstruct GmbH, Nuremberg, Germany; and Olivier Bockenbach, Mercury Computer Systems, Berlin, Germany

Tomographic image reconstruction is computationally very demanding. In all cases the backprojection represents the performance bottleneck due to the high operational count and the high demand placed on the memory subsystem. In the past, solving this problem has led to the use of Digital Signal Processors and the implementation of specific architectures, connecting Application Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs) to the memory through dedicated high-speed busses. More recently, attempts have been made to use Graphic Processing Units (GPUs) and the Cell Broadband Engine processor (CBE).

It is worth noting that, however much these architectures differ, they all share common properties that make them attractive for the implementation of backprojection algorithms: the relative balance of high memory bandwidth and processing capabilities.

The backprojection is a considerable processing step, but a 3D image reconstruction algorithm also includes pre-processing, filtering and post-processing steps. Various devices considered for accelerating the backprojection reveal themselves to be more or less usable for other tasks, requiring more or less assistance from other co-processing units. The implementation of the complete reconstruction pipeline may require the combination of several different devices, influencing the design, implementation and maintenance costs.

Methods and Material Hardware

Four different platforms were selected:

  • Our reference platform for image quality is on a standard PC with a single Xeon processor clocked at 3.06 GHz and a front bus side with 533 MHz
  • The PCI Express Cell Accelerator Board from Mercury Computer Systems
  • The PCI VantageRT-FCN board from Mercury Computer Systems
  • The G70 GPU from NVIDIA.


1 2 3 4 5 6 next

Subscribe to our RSS Feeds