Comprehensive Analysis of High-Performance Computing Methods for Filtered Back-Projection

• Mendl, Christian B. ^[1] ; Eliuk, Steven N. ^[2] ; Noga, Michelle ^[2] ; Boulanger, Pierre ^[2]
  1. [1] Technical University Munich

     Technical University Munich

     Kreisfreie Stadt München, Alemania

     
  2. [2] University of Alberta

     University of Alberta

     Canadá

     
• Localización: ELCVIA. Electronic letters on computer vision and image analysis, ISSN-e 1577-5097, Vol. 12, Nº. 1, 2013 (Ejemplar dedicado a: December 2013), págs. 1-16
• Idioma: inglés
• Enlaces
  • Texto completo  1    2  
• Resumen

  • This paper provides an extensive runtime, accuracy, and noise analysis of Computed Tomography (CT) reconstruction algorithms using various High-Performance Computing (HPC) frameworks such as: “conventional” multi-core, multi threaded CPUs, Compute Unified Device Architecture (CUDA), and DirectX or OpenGL graphics pipeline programming. The proposed algorithms exploit various built-in hardwired features of GPUs such as rasterization and texture filtering. We compare implementations of the Filtered Back-Projection (FBP) algorithm with fan-beam geometry for all frameworks. The accuracy of the reconstruction is validated using an ACR-accredited phantom, with the raw attenuation data acquired by a clinical CT scanner. Our analysis shows that a single GPU can run a FBP reconstruction 23 time faster than a 64-core multi-threaded CPU machine for an image of 1024 × 1024. Moreover, directly programming the graphics pipeline using DirectX or OpenGL can further increases the performance compared to a CUDA implementation.