Computer Science and Engineering Department Seminar

Visiting Speaker: Dr. Klaus Mueller
Time: 2pm-3pm
Date: Wednesday February 20th 2008
Place:  UH-341

Title: Real-time 3D computed tomographic reconstruction using commodity 
graphics hardware

Dr. Mueller is an Associate Professor of Computer Science and head of the
Computational Medicine Group within the Center for Visual Computing at
Stony Brook University. He has published more than 90 conference and
journal papers on medical imaging, visualization, and general-purpose
GPU-accelerated computing. He has been actively involved in
GPU-accelerated CT from the early days of SGI texture mapping workstations
to today's PC-based commodity boards. His journal articles in IEEE
Transactions on Medical Imaging and Nuclear Science have documented these
pioneering efforts and have been widely referenced. He teaches courses on
medical imaging, GPU-based computing, and computational medicine at both
the university and international symposium levels. He is currently
co-funded by an NIH grant focused on this topic. He is a senior member of
the IEEE.

Abstract
The recent emergence of various types of flat-panel x-ray detectors and
C-arm gantries now enables the construction of novel imaging platforms
for a wide variety of clinical applications. Many of these applications
require interactive 3D image generation, which cannot be satisfied with
inexpensive PC-based solutions using the CPU.We present a solution based
on commodity graphics hardware (GPUs) to provide these capabilities.
While GPUs have been employed for CT reconstruction before, our approach
provides significant speedups by exploiting the various built-in
hardwired graphics pipeline components for the most expensive CT
reconstruction task, backprojection. The result is a novel streaming CT
framework that conceptualizes the reconstruction process as a steady  flow
of data across a computing pipeline, updating the reconstruction  result
immediately after the projections have been acquired. Using a  single PC
equipped with a single high-end commodity graphics board (the  Nvidia 8800
GTX), our system is able to process clinically-sized
projection data at speeds meeting and exceeding the typical flat-panel
detector data production rates, enabling throughput rates of 40-50
projections/s for the reconstruction of 512^3 volumes. Finally, I will
also present results on GPU-accelerating iterative CT reconstruction
algorithms, some applications for these, and summarize our work in the
areas of GPU-accelerated volume visualization and scientific simulation.
>