Graphic processing unit-accelerated mutual information-based 3D image rigid registration

Abstract

Mutual information (MI)-based image registration is effective in registering medical images, but it is computationally expensive. This paper accelerates MI-based image registration by dividing computation of mutual information into spatial transformation and histogram-based calculation, and performing 3D spatial transformation and trilinear interpolation on graphic processing unit (GPU). The 3D floating image is downloaded to GPU as flat 3D texture, and then fetched and interpolated for each new voxel location in fragment shader. The transformed results are rendered to textures by using frame buffer object (FBO) extension, and then read to the main memory used for the remaining computation on CPU. Experimental results show that GPU-accelerated method can achieve speedup about an order of magnitude with better registration result compared with the software implementation on a single-core CPU.

Keywords

image registration / mutual information / graphic processing unit (GPU)

References

[1] Viola P, Wells W M. Alignment by maximization of mutual information[C]. In: Proceedings of the 5th International Conference on Computer Vision. Boston, 1995. 16–23.

[2] Pluim J. P. W., Maintz J. B. A., Viergever M. A. Mutualinformation-based registration of medical images: A survey[J]. IEEE Transactions on Medical Imaging, 2003, 22(8): 986-1004.

[3] Zhang H. Y., Zhang J. W., Sun J. Z. Registration method for CT-MR image based on mutual information[J]. Transactions of Tianjin University, 2007, 13(3): 226-230.

[4] Jin R. C., Wang J. H., Song E. M. An algorithm for brain MR image registration based on rough registration and mutual information[J]. Computer Simulation, 2007, 24(4): 61-63.

[5] Pluim J. P. W., Maintz J. B. A., Viergever M. A. Mutual information matching in multiresolution contexts[J]. Image and Vision Computing, 2001, 19(1): 45-52.

[6] Maes F., Vandermeulen D., Suetens P. Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual information[J]. Medical Image Analysis, 1999, 3(4): 373-386.

[7] Rohlfing T., Maurer C. R. Jr. Nonrigid image registration in shared-memory multiprocessor environments with application to brains, breasts, and bees[J]. IEEE Transactions on Information Technology in Biomedicine, 2003, 7(1): 16-25.

[8] Carlos R. C., Jogikal M. J., Raj S. FAIR: A hardware architecture for real-time 3D image registration[J]. IEEE Transactions on Information Technology in Biomedicine, 2003, 7(4): 426-434.

[9] Moriyoshi O, Hangu Y, Frank S et al. Accelerating mutual-information-based linear registration on the cell broadband engine processor[C]. In: IEEE International Conference on Multimedia and Expo. Beijing, China, 2007. 272–275.

[10] Shams R, Barnes N. Speeding up mutual information computation using NVIDIA CUDA hardware[C]. In: Proceedings of the 9th Biennial Conference of the Australian Pattern Recognition Society on Digital Image Computing Techniques and Applications. Glenelg, Australia, 2007. 555–560.

[11] Lin Yuping, Medioni G. Mutual information computation and maximization using GPU[C]. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Alaska, USA, 2008. 1–6.

[12] Strzodka R., Droske M., Rumpf M. Image registration by a regularized gradient flow — A streaming implementation in DX9 graphics hardware[J]. Computing, 2004, 73(4): 373-389.

[13] Strzodka R., Droske M., Rumpf M. Fast image registration in DX9 graphics hardware[J]. Journal of Medical Informatics and Technologies, 2003, 6 43-49.

[14] Harris M J, Baxter W V, Scheuermann T et al. Simulation of cloud dynamics on graphics hardware[C]. In: Proceedings of the SIGGRAPH/Eurographics Workshop on Graphics Hardware. San Diego, California, USA. 2003. 92–101.

[15] Alexander K., Johann D., Felix R., et al. GPU accelerated image registration in two and three dimensions[C] Proceedings of Bildverarbeitung für die Medizin, 2006, Berlin, Heidelberg: Springer 261-265.