High performance model based image reconstruction

In Computed Tomography (CT) methods, Model Based Iterative Reconstruction (MBIR) produces higher quality images than commonly used Filtered Backprojection (FBP) but at a very high computational cost. We describe a new MBIR implementation, PSV-ICD, which significantly reduces the computational cost of MBIR while retaining its benefits. It describes a novel organization of the scanner data into super-voxels (SV) that, combined with a super-voxel buffer (SVB), dramatically increases locality and prefetching while enabling parallelism across SVs. Experimental data is presented showing an average speedup of 187X on 20 cores.

• Type: Journal
• Year: 2016
• Where: PPOPP
• Authors: Xiao Wang, Amit Sabne, Sherman J. Kisner, Anand Raghunathan, Charles A. Bouman, Samuel P. Midkiff
• Keywords: Distributed And Parallel Computing, PPOPP 2016