Parallel Implementation of Bias Field Correction Fuzzy C-Means Algorithm for Image Segmentation

Image segmentation in the medical field is one of the most important phases to diseases diagnosis. The bias field estimation algorithm is the most interesting techniques to correct the in-homogeneity intensity artifact on the image. However, the use of such technique requires a powerful processing and quite expensive for big size as medical images. Hence the idea of parallelism becomes increasingly required. Several researchers have followed this path mainly in the bioinformatics field where they have suggested different algorithms implementations. In this paper, a novel Single Instruction Multiple Data (SIMD) architecture for bias field estimation and image segmentation algorithm is proposed. In order to accelerate compute-intensive portions of the sequential implementation, we have implemented this algorithm on three different graphics processing units (GPU) cards named GT740m, GTX760 and GTX580 respectively, using Compute Unified Device Architecture (CUDA) software programming tool. Numerical obtained results for the computation speed up, allowed us to conclude on the suitable GPU architecture for this kind of applications and closest ones.

[1]  J. Bezdek,et al.  FCM: The fuzzy c-means clustering algorithm , 1984 .

[2]  Michael Unser,et al.  Multigrid adaptive image processing , 1995, Proceedings., International Conference on Image Processing.

[3]  Jagath C. Rajapakse,et al.  Statistical approach to segmentation of single-channel cerebral MR images , 1997, IEEE Transactions on Medical Imaging.

[4]  Aly A. Farag,et al.  A modified fuzzy c-means algorithm for bias field estimation and segmentation of MRI data , 2002, IEEE Transactions on Medical Imaging.

[5]  Jan Sijbers,et al.  Bias Field Correction for MRI Images , 2005, CORES.

[6]  Chris Harris,et al.  Iterative Solutions using Programmable Graphics Processing Units , 2005, FUZZ-IEEE.

[7]  Bostjan Likar,et al.  A Review of Methods for Correction of Intensity Inhomogeneity in MRI , 2007, IEEE Transactions on Medical Imaging.

[8]  James M. Keller,et al.  Incorporation of Non-euclidean Distance Metrics into Fuzzy Clustering on Graphics Processing Units , 2007, Analysis and Design of Intelligent Systems using Soft Computing Techniques.

[9]  Derek T. Anderson,et al.  Parallelisation of fuzzy inference on a graphics processor unit using the compute nified device architecture , 2008 .

[10]  Liang Liao,et al.  MRI brain image segmentation and bias field correction based on fast spatially constrained kernel clustering approach , 2008, Pattern Recognit. Lett..

[11]  Manoranjan Dash,et al.  Graphics Hardware based Efficient and Scalable Fuzzy C-Means Clustering , 2008, AusDM.

[12]  James M. Keller,et al.  Speedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units , 2008, IEEE Transactions on Fuzzy Systems.

[13]  Saraju P. Mohanty.  GPU-CPU multi-core for real-time signal processing , 2009, 2009 Digest of Technical Papers International Conference on Consumer Electronics.

[14]  Kang Zhang,et al.  Real-time 4D signal processing and visualization using graphics processing unit on a regular nonlinear-k Fourier-domain OCT system , 2010, Optics express.

[15]  S. R. Kannan,et al.  Modified Bias Field Fuzzy C-Means for Effective Segmentation of Brain MRI , 2010, Trans. Comput. Sci..

[16]  Zexuan Ji,et al.  A modified possibilistic fuzzy c-means clustering algorithm for bias field estimation and segmentation of brain MR image , 2011, Comput. Medical Imaging Graph..

[17]  O. Nalcioglu,et al.  A new bias field correction method combining N3 and FCM for improved segmentation of breast density on MRI. , 2010, Medical physics.

[18]  Chunming Li,et al.  A Level Set Method for Image Segmentation in the Presence of Intensity Inhomogeneities With Application to MRI , 2011, IEEE Transactions on Image Processing.

[19]  Lei Xing,et al.  GPU computing in medical physics: a review. , 2011, Medical physics.

[20]  S. Hariri,et al.  Exploiting GPUs for compute-intensive medical applications , 2012, 2012 International Conference on Multimedia Computing and Systems.

[21]  Jarosław Gocławski,et al.  Cuda Based Fuzzy C-Means Acceleration for the Segmentation of Images with Fungus Grown in Foam Matrices , 2012 .

[22]  Mohamed Batouche,et al.  ACPSO: A Novel Swarm Automatic Clustering Algorithm Based Image Segmentation , 2012 .

[23]  Lei Wang,et al.  Fast Implementations of the Levelset Segmentation Method With Bias Field Correction in MR Images: Full Domain and Mask-Based Versions , 2013, IbPRIA.

[24]  H. Ding,et al.  Breast density quantification using magnetic resonance imaging (MRI) with bias field correction: a postmortem study. , 2013, Medical physics.

[25]  Anders Eklund,et al.  Medical image processing on the GPU - Past, present and future , 2013, Medical Image Anal..

[26]  Devrim Akgün,et al.  GPU-accelerated dynamic functional connectivity analysis for functional MRI data using OpenCL , 2014, EIT.

[27]  Haiyang Li,et al.  An Improved Image Segmentation Algorithm Based on GPU Parallel Computing , 2014, J. Softw..

[28]  Mahmoud Al-Ayyoub,et al.  A GPU-based implementations of the fuzzy C-means algorithms for medical image segmentation , 2015, The Journal of Supercomputing.

[29]  Frank Lindseth,et al.  Medical image segmentation on GPUs - A comprehensive review , 2015, Medical Image Anal..