Enhancements to Reconstruction Techniques in Computed Tomography Using High Performance Computing

  • Author / Creator
    Eliuk, Steven N
  • Computers have been used in diagnostic imaging for decades, but High-Performance Computing (HPC) in diagnostic imaging is rather rare because of the high cost associated with traditional HPC. Recent advancements in shared memory computers and the large market for commodity graphic cards have provided the means for technological evolution at a more rapid pace and decreased cost. These improvements can provide a detailed 3D view of the human body, near instantaneous recon- struction for Computed Tomography (CT), more intricate reconstruction algorithms, and reduced radiation exposure to patients through the use of iterative reconstruction techniques. The use of HPC in diagnostic imaging is realizable with current commodity hardware. For this reason a new reconstruction, visualization, and interaction environment has been constructed to decrease recon- struction time, decrease radiation exposure, and reconstruct images from raw attenuation values in medical-CT. The thesis explores the use of different computational tools in the Advanced Recon- struction Environment for Medical Imaging (AREMI). AREMI uses multi-core and multi-GPU pro- gramming for reconstruction of real-world clinical raw attenuation values from Siemens Definition Flash CT scanners. The use of this environment, analysis of various reconstruction techniques, and the ability to output High Dynamic Range (HDR) has provided a means to analyze various aspects of CT reconstruction in medical applications. The thesis also provides a concise review of filtered back-projection and algebraic methods in CT-reconstruction and provides insight into programming advanced CT-reconstruction concepts in a serial, multi-CPU, GPU, and multi-GPU environment that are not formally explained in literature; therefore, making this topic accessible and valuable to a broad audience.

  • Subjects / Keywords
  • Graduation date
    Fall 2012
  • Type of Item
  • Degree
    Doctor of Philosophy
  • DOI
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.
  • Language
  • Institution
    University of Alberta
  • Degree level
  • Department
  • Specialization
    • Computing Science
  • Supervisor / co-supervisor and their department(s)
  • Examining committee members and their departments
    • Lu, Paul (Computing Science)
    • Hoover, James (Computing Science)
    • Noga, Michelle (Radiology and Diagnostic Imaging)
    • Boulanger, Pierre (Computing Science)
    • Beg, Faisal (Biomedical Engineering, Simon Fraser University)