论文信息 - Tetrahedral Interpolation for Deformable Image Registration on GPUs

Tetrahedral Interpolation for Deformable Image Registration on GPUs

We speed up the tetrahedral interpolation step of a deformable image registration code called MORFEUS. We implement several versions of the interpolation code on a Fermi GPU (GTX480). Despite the irregularity of the code, we obtained kernel speedups of up to 24.6x, 33.7x and 62.4x on three real-lif e benchmarks. These numbers do not include the data transfer time between the CPU and the GPU because it can be amortized over the other steps of MORFEUS code. However, we do explore the impact of different GPU data transfer techniques.

Tarek S. Abdelrahman | Cedomir Segulja | Tianyi David Han | Kristy K. Brock | Joanne Moseley

[1] Balraj Naren,et al. Medical Image Registration , 2022 .

[2] John D. Owens,et al. Fast Deformable Registration on the GPU: A CUDA Implementation of Demons , 2008, 2008 International Conference on Computational Sciences and Its Applications.

[3] K. Brock,et al. Accuracy of finite element model-based multi-organ deformable image registration. , 2005, Medical physics.

[4] Heinz-Otto Peitgen,et al. GPU Accelerated Image Registration in Two and Three Dimensions , 2006, Bildverarbeitung für die Medizin.

[5] Bo Li,et al. GPU Accelerated Non-rigid Registration for the Evaluation of Cardiac Function , 2008, MICCAI.

[6] Max A. Viergever,et al. A survey of medical image registration , 1998, Medical Image Anal..

[7] Michael Kaus,et al. Deformable Image Registration for Radiation Therapy Planning: Algorithms and Applications , 2007 .