Clique Identification and Propagation for Multimodal Brain Tumor Image Segmentation

Brain tumors vary considerably in size, morphology, and location across patients, thus pose great challenge in automated brain tumor segmentation methods. Inspired by the concept of clique in graph theory, we present a clique-based method for multimodal brain tumor segmentation that considers a brain tumor image as a graph and automatically segment it into different sub-structures based on the clique homogeneity. Our proposed method has three steps, neighborhood construction, clique identification, and clique propagation. We constructed the neighborhood of each pixel based on its similarities to the surrounding pixels, and then extracted all cliques with a certain size k to evaluate the correlations among different pixels. The connections among all cliques were represented as a transition matrix, and a clique propagation method was developed to group the cliques into different regions. This method is also designed to accommodate multimodal features, as multimodal neuroimaging data is widely used in mapping the tumor-induced changes in the brain. To evaluate this method, we conduct the segmentation experiments on the publicly available Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) dataset. The qualitative and quantitative results demonstrate that our proposed clique-based method achieved better performance compared to the conventional pixel-based methods.

[1]  Stefan Bauer,et al.  Fully Automatic Segmentation of Brain Tumor Images Using Support Vector Machine Classification in Combination with Hierarchical Conditional Random Field Regularization , 2011, MICCAI.

[2]  Nikos Paragios,et al.  Graph-based detection, segmentation & characterization of brain tumors , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Lei Jing,et al.  Localized multiscale texture based retrieval of neurological image , 2010, 2010 IEEE 23rd International Symposium on Computer-Based Medical Systems (CBMS).

[4]  E. Holland.  Progenitor cells and glioma formation , 2001, Current opinion in neurology.

[5]  Kathryn Fraughnaugh,et al.  Introduction to graph theory , 1973, Mathematical Gazette.

[6]  Alan L. Yuille,et al.  Efficient Multilevel Brain Tumor Segmentation With Integrated Bayesian Model Classification , 2008, IEEE Transactions on Medical Imaging.

[7]  Leo Joskowicz,et al.  Automatic segmentation, internal classification, and follow-up of optic pathway gliomas in MRI , 2012, Medical Image Anal..

[8]  Sidong Liu,et al.  A 3D difference-of-Gaussian-based lesion detector for brain PET , 2012, 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI).

[9]  Olivier Clatz,et al.  Glioma Dynamics and Computational Models: A Review of Segmentation, Registration, and In Silico Growth Algorithms and their Clinical Applications , 2007 .

[10]  Sidong Liu,et al.  Multiscale and multiorientation feature extraction with degenerative patterns for 3D neuroimaging retrieval , 2012, 2012 19th IEEE International Conference on Image Processing.

[11]  Stan Z. Li,et al.  Markov Random Field Modeling in Image Analysis , 2001, Computer Science Workbench.

[12]  David Dagan Feng,et al.  Beating cilia identification in fluorescence microscope images for accurate CBF measurement , 2015, 2015 IEEE International Conference on Image Processing (ICIP).

[13]  Gustavo Carneiro,et al.  A Discriminative Model-Constrained Graph Cuts Approach to Fully Automated Pediatric Brain Tumor Segmentation in 3-D MRI , 2008, MICCAI.

[14]  S. Bauer,et al.  A survey of MRI-based medical image analysis for brain tumor studies , 2013, Physics in medicine and biology.

[15]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[16]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[17]  Jerry L Prince,et al.  Current methods in medical image segmentation. , 2000, Annual review of biomedical engineering.

[18]  Sidong Liu,et al.  Localized functional neuroimaging retrieval using 3D discrete curvelet transform , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[19]  D. Louis Collins,et al.  Hierarchical Probabilistic Gabor and MRF Segmentation of Brain Tumours in MRI Volumes , 2013, MICCAI.

[20]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[21]  Leif H. Finkel,et al.  CURRENT METHODS IN MEDICAL IMAGE SEGMENTATION1 , 2007 .

[22]  Sidong Liu,et al.  A robust volumetric feature extraction approach for 3D neuroimaging retrieval , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[23]  W. Eric L. Grimson,et al.  Anatomical guided segmentation with non-stationary tissue class distributions in an expectation-maximization framework , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[24]  W. Eric L. Grimson,et al.  A Unifying Approach to Registration, Segmentation, and Intensity Correction , 2005, MICCAI.

[25]  Sidong Liu,et al.  Multimodal neuroimaging computing: the workflows, methods, and platforms , 2015, Brain Informatics.

[26]  Sidong Liu,et al.  Multimodal neuroimaging computing: a review of the applications in neuropsychiatric disorders , 2015, Brain Informatics.

[27]  Sidong Liu,et al.  Hierarchical and binary spatial descriptors for lung nodule image retrieval , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.