FPGA-based forward and back-projection operators for tomographic reconstruction

Kyungchan Jin; Sangyup Song

doi:10.1117/12.2007533

6 March 2013 FPGA-based forward and back-projection operators for tomographic reconstruction

Kyungchan Jin, Sangyup Song

Proceedings Volume 8668, Medical Imaging 2013: Physics of Medical Imaging; 866836 (2013) https://doi.org/10.1117/12.2007533
Event: SPIE Medical Imaging, 2013, Lake Buena Vista (Orlando Area), Florida, United States

Abstract

Artifacts in computerized tomography (CT), such as metal streak effects, can also be reduced using the iterative reconstruction approach. The main issue in iterative method is the computation cost by efficient implementation of the forward and back-projection operations, which are the dominant cost in all iterative reconstruction algorithms. We designed a field programmable gate array (FPGA)-based operators for iterative forward and back-projection method to solve the artifact model. The projection method in CT reconstruction is to retrieve the volumetric image based on observed projection image and makes reconstruction errors minimize. The FPGA-based operators contain only the iterative reconstruction operations from tomographic projections, and the filtering of detector data and the geometry correction between detector and object are done by host CPU processor. For FPGA design, we used Impulse C package, C-to-FPGA tool including the use of streaming and pipelining for high performance. We evaluated the FPGA-based projection on Shepp-Logan phantom data with metal streak artifact. Simulation results show that the FPGA-based operators can reduce the computation time of iterative reconstruction, while still providing accuracy comparable to CPU or GPU-based reconstruction.

Citation Download Citation

Kyungchan Jin and Sangyup Song "FPGA-based forward and back-projection operators for tomographic reconstruction", Proc. SPIE 8668, Medical Imaging 2013: Physics of Medical Imaging, 866836 (6 March 2013); https://doi.org/10.1117/12.2007533

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available