GPU based Gaussian Bilateral Filter for Noisy MR Image

S.D. Oza¹ , K.R. Joshi²

Section:Research Paper, Product Type: Journal-Paper
Vol.8 , Issue.3 , pp.39-44, Jun-2020

Online published on Jun 30, 2020

Copyright © S.D. Oza, K.R. Joshi . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 

View this paper at   Google Scholar | DPI Digital Library

XML View     PDF Download

How to Cite this Paper

Abstract :
Magnetic resonance imaging (MRI) diagnostic technique is significantly used to diagnose disorders and deformities of the knee. The process of MRI corrupts the MR image with signal dependent rician noise which hinders accuracy of diagnosis. The paper proposes a new Gaussian Bilateral Filter giving better performance as against bilateral only and iterative bilateral filter, especially for knee MRI with high noise level. The research work further provides an accelerated solution to the challenge faced by the proposed filter for large resolution images. The work here implements the proposed Gaussian Bilateral filter using CUDA GPU there by achieving real time performance for knee MR image with 64Mega pixel resolution. The speed up achieved with CUDA GPU implementation of the proposed filter is 320.064 times more than its CPU counterpart. This would surely contribute to intelligent diagnostic setups

Key-Words / Index Term :
Gaussian Bilateral Filter, MRI, CUDA GPU, Shared memory

References :
[1] J. Hornak ?The Basics of MRI?, e-book, https://www.cis.rit.edu/htbooks/mri
[2] K R Joshi And Oza S D, ?Mri Denoising For Healthcare?, M. H. Kolekar And Vinod Kumar (Ed) In Biomedical Signal And Image Processing In Patient Care, Igi Global, pp 65-85, 2017
[3] H. Gudbjartsson. And S. Partz, ?The Rician distribution of noisy MRI data?, in HHS Public Access Manuscript, PMC, 2008
[4] J. Mohan, V. Krishnaveni and Y. Guo, ?A survey on magnetic resonance image denoising methods?, in Biomedical Signal processing and Control, Elsevier, 9( 1), pp 56-59, 2014
[5] S. D. Oza and K. R. Joshi, ?Performance Analysis of Denoising Filters for MR Images? Amlan Chakrabarti, Neha Sharma,Valentina Emilia Balas (Eds) In Advances in Computing Applications, Springer Singapore, pp 87-96, 2016
[6] G. Gerig, O. Kubler and F. Jolez, ?Nonlinear anisotropic filtering of MRI data?, in IEEE Transactions. Medical aging, 11(2), pp. 221?232, 1992
[7] P. Coupe et al, ?Fast non local means denoising for MR images?, in : Proceedings at the 9th International Conference on Medical Image Computing and Computer assisted Intervention (MICCAI), Copenhagen, pp. 33?40, 2006.
[8] P. Coupe et al., ?An optimized block wise nonlocal means denoising filter for 3-D magnetic resonance images?, IEEE Trans. Med. Imaging, 27(4), pp. 425?441, 2008
[9] P. Coupe et al, ?3D wavelet sub bands mixing for image denoising?, Int. J. Biomed. Imaging , http://dx.doi.org/10.1155/2008/590183, Article ID 590183, 2008.
[10] Liu et al., ?Noise Estimation from Single Image?, proc. IEEE Computer Vision and Pattern Recognition, 1, pp. 901-908, 2006
[11] H. Liu, Y. Cihui, P. Ning, S. Enmin, G. Richard G., ?Denoising 3D MR images by the enhanced non-local means filter for rician noise?, in Magnetic Resonance Imaging, 28(10) pp. 1485?1496, 2010
[12] J. Manj?n, J. Carbonell-Caballero, J. Lull, G. Garc?a-Mart?, L. Mart?-Bonmat?. and M. Robles, ?MRI denoising using non-local means?, in Medical Image Anaysis., Elsevier, 12(4), pp. 514?523, 2008.
[13] J. Manj?n, P. Coup?, L. Mart?-Bonmat?, D. Collins, M. Robles, ?Adaptive non-local means denoising of MR images with spatially varying noise levels?, in Magnetic Resonance Imaging, 31(1), pp. 192?203, 2010
[14] N. Wiest-Daessl? et al , ?Rician noise removal by non-local means filtering for low signal-to-noise ratio MRI: applications to DT-MRI?, Med. Image Comput. Comput. Assist. Interv. 11(2), pp. 171?179, 2008.
[15] M. Elad, ?On the Bilateral Filter and Ways To Improve It,? IEEE transaction on Image processing, 11(10), pp. 1141-1151, 2002
[16] W. Wong et al, ?Trilateral Filtering for Biomedical Images,? in Proc. IEEE Int. Symposium Biomedical Imaging, pp. 820-823, 2004
[17] R. Riji, ?Iterative Bilateral Filter for Rician Noise Reduction in MR Images,? Signal, Image and Video Processing, 9(7), pp 1543-1548, 2015
[18] V. Aurich and J. Weule, ?Nonlinear Gaussian Filters Performing Edge preserving Diffusion?, In Mutererkennung, 17, DAGM-Symposium, pp 538-545, Springer-verlog, 1995
[19] J. Larsson, ?A Case Study of Parallel Bilateral Filtering on the GPU?, Master Thesis, Malardalen University 2015
[20] M. Howison, ?Comparing GPU Implementations of Bilateral and Anisotropic Diffusion Filters for 3D Biomedical Datasets?, Computational Research Division, Lawrence Berkeley National Laboratory, USA, 2018
[21] M. Howison, and E. W. Bethel, ?GPU-accelerated denoising of 3D magnetic resonance Images?, Journal of Real-Time Image Processing (JRTIP) Springer, vol. 13, issue 4, pp. 713-724, 2017
[22] J. Fengjiao, S. Dan and C. Liu, Dong. ?Fast Adaptive Ultrasound Speckle Reduction with Bilateral Filter on CUDA?, In Proc. of 5th Int Conf on Bioinformatics and Biomedical Engineering, ICBBE, pp. 1-4, 2011
[23] S. Smith and J. Brady, ?Susan ? a new approach to low level image processing?, Int. Journal Filtering for Gray and Color Images?, in Proc. Int. Conf Computer Vision Vol. 23, pp. 45-78, 1997
[24] Tomasi C and Manduchi R, ?Bilateral. Computer Vision, pp. 839-846, 1998
[25] Paris S et al, ?Filtering: Theory and Applications?, Computer Graphics and Vision,. 4(1), pp. 1-73, 2008
[26] Nvidia CUDA C programming Guide version 4.2, 2012
[27] CUDA C Best Practices Guide DG-05603-001_v5.0, Design Guide, October 2012
[28] D. Granata, U. Amato, B. Alfano, MRI Denoising by non local means on multi-GPU, Journal of Real Time Image Processing, ACM, 16(2), 2019
[29] P K Jithy,, P Simon, ?Review on Automatic Segmentation Techniques in Medical Images?, International Journal of Scientific Research in Computer Science, Engineering and Information Technology, IJSRCSEIT, 2(4,) ISSN : 24356-3307, 2017