基于空间FCM与MRF方法的乳腺MRI序列三维病灶分割研究

doi:10.3969/j.issn.0258-821. 2014. 02.010

摘要
图/表
参考文献
相关文章 (0)

全文: PDF (2748 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对乳腺DCE-MRI病灶分割，提出一种空间FCM聚类与MRF随机场相结合的三维分割方法。首先，对MRI图像进行空间FCM粗分割，提取病灶粗轮廓。然后，在其基础上进行MRF精分割，并结合病灶三维信息：用相邻切片分割结果对应标号矩阵初始化MRF精分割标号场，同时用该张切片粗分割所得隶属度矩阵对MRF精分割进行参数自适应调整。用该方法与空间FCM、水平集、模糊MRF方法对50例MRI数据进行分割对比实验，得到良、恶性病灶分割重叠率分别为76.4、75.5；相比于空间FCM的68.%、69.5水平集的70.8、72.6以及模糊MRF的72.9、73.6有明显提升。对所有175例MRI数据分割结果进行非监督评价，得到良、恶性病灶区域均匀性均大于0.92；区域内差异性良性病灶92%小于150、恶性病灶98%小于150；区域间差异性良性病灶87%大于0.25、恶性病灶90%大于0.3综上表明，该方法具有较高的分割精度。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张承杰力华^*

关键词 ：乳腺DCE-MRI图像, 灶分割, CM-MRF, 维分割, 数自适应

Abstract：Breast MRI image segmentation is a challenging issue. This paper presents a 3D segmentation method, which is based on spatial FCM clustering and Markov random field. First, the MRI image was coarse segmented by spatial FCM to extract lesion contours. And then MRF segmentation was conducted to refine the result. We combined the 3D information of lesion in the MRF process by using segmentation result of contiguous slices to constraint the slice segmentation. At the same time, a membership matrix of FCM segmentation result is used for adaptive adjustment of Markov parameters in MRF segmentation process. The segmentation performance of this method was compared with that of spatial FCM, level set and fuzzy MRF on a database including 50 breast DCE-MRI examinations. Results show that average overlap rate for benign and malignant of our method is 76.4% and 75.5% respectively, compared with
68.7% and 69.5% for spatial FCM, 70.8% and 72.6% for level set method,and 72.9% and 73.6% for fuzzy MRF. It is demonstrated that our method has a better performance in accuracy. In addition, we used unsupervised evaluation method to evaluate the segmentation result of all the 175 breast DCE-MRI image sequences in the database, The uniformities of intra region (URs) for both benign and malignant lesion was more than 0.92. The differences within region (DRs) of 92% of the benign lesions and 98% of the malignant lesions were less than 150, the differences of the inter region (DIR) of 87% of benign lesion were more than 0.25, while those of 90% of malignant lesion were over 0.3. The results demonstrated the proposed method has a good performance in segmentation accuracy.

Key words： breast DCE-MRI image lsion segmentation FCM-RF 3D segmentation parameter adaptive

基金资助:国家自然科学基金（61271063）；国家重点基础研究发展计划资助(2013CB329502)；国家杰出青年科学基金（60788101）

引用本文:

张承杰力华^*. 基于空间FCM与MRF方法的乳腺MRI序列三维病灶分割研究[J]. 中国生物医学工程学报, 2014, 33(2): 202-211.
ZHANG ChengJie LiHua^*. 3D Segmentation of Breast DCE-MRI Sequence Using Spatial FCM-MRF Method. journal1, 2014, 33(2): 202-211.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-821. 2014. 02.010 或 http://cjbme.csbme.org/CN/Y2014/V33/I2/202

［1］Jemal A, Siegel R, Xu J, et al. Cancer statistics, 2010［J］. A Cancer Journal for Clinicians, 2010, 60(5): 277-300.
［2］Smith R A, Cokkindes V, Brooks D, et al. Cancer screening in the United States, 2011［J］. A Cancer Journal for Clinicians, 2011, 61(1): 8-30.
［3］Saslow D, Boetes C, Burke W, et al. American cancer society guidelines for breast screening with MRI as an adjunct to mammography ［J］. A Cancer Journal for Clinicians, 2007, 57(2): 75-89. 
［4］Lucas-Quesada FA, Sinha U, Sinha S. Segmentation strategies for breast tumors from dynamic MR images ［J］. Journal of Magnetic Resonance Imaging, 1996, 6(5): 753-763.［5］Gilhuijs KGA, Giger ML, Bick U. A method for computerized assessment of tumor extent in contrastenhanced MR images of the breast ［C］//Doi K, MacMahon H, Giger M L，et al,eds. ComputerAided Diagnosis in Medical Imaging. Chicago: Elsevier Science, 1999: 305-310. 
［6］Chen Weijie, Giger ML, Bick U. A fuzzy cmeans (FCM)-based approach for computerized segmentation of breast lesions in dynamic contrastenhanced MR images ［J］. Academic Radiology, 2006, 13(1): 63-72. 
［7］Wu Qiu, Salganicoff M, Krishnan A, et al. Interactive lesion segmentation on dynamic contrast enhanced breast MRI using a Markov model ［C］// Joseph M. Reinhardt, Josien P. W. Pluim eds. International Society for Optics and Photonics. San Diego: SPIE, 2006: 61444M-61444M-8. 
［8］Szabó BK, Aspelin P, Wiberg MK. Neural network approach to the segmentation and classification of dynamic magnetic resonance images of the breast: comparison with empiric and quantitative kinetic parameters ［J］. Academic Radiology, 2004, 11(12): 1344-1352.
［9］Zhang Kaihua, Zhang Lei, Song Huihui, et al. Active contours with selective local or global segmentation: A new formulation and level set method ［J］. Image and Vision Computing, 2010, 28(4): 668-676.
［10］Agner SC, Xu Jun, Madabhushi A. Spectral embedding based active contour (SEAC) for lesion segmentation on breast dynamic contrast enhanced magnetic resonance imaging ［J］. Medical Physics, 2013, 40: 032305-1- 032305-12.
［11］Shi J, Sahiner B, Chan HP, et al. Breast mass segmentation on dynamic contrastenhanced magnetic resonance scans using the level set method ［C］//Maryellen L. Geger, Nico karssemeijer, eds. International Society for Optics and Photonics. San Diego: SPIE, 2008: 69152A. 1-69152A. 7. 
［12］Huang C L, Kuo L Y. ShapeBased Level Set Method for Breast Mass Segmentation on Breast MRI［M］//Advances in Computer Science and Education Applications. Berlin：SpringerHeidelberg, 2011: 373-379. 
［13］Bahreini L, Fatemizadeh E, Gity M. Gradient vector flow snake segmentation of breast lesions in Dynamic ContrastEnhanced MR images［C］//Biomedical Engineering (ICBME). Isfahan: IEEE, 2010: 1-4. 
［14］Cai Weiling, Song Chen, Zhang Daoqiang. Fast and robust fuzzy cmeans clustering algorithms incorporating local information for image segmentation ［J］. Pattern Recognition, 2007, 40(3): 825-838.
［15］徐荣．马尔科夫随机场理论在图像分割领域的应用［D］．南京：东南大学，2009．
［16］张石, 董建威, 佘黎煌. 医学图像分割算法的评价方法［J］. 中国图像图形学报. 2009, 14(9): 1872-1880.
［17］冯前进, 陈武凡. 模糊马尔科夫模型与图像分割新算法［J］. 南方医科大学学报, 2006, 26(5): 579-583.
［18］Li BN, Chui CK, Chang S, et al. Integrating spatial fuzzy clustering with level set methods for automated medical image segmentation ［J］. Computers in Biology and Medicine, 2011, 41(1): 1-10.