北京生物医学工程

___________图论在脑肿瘤分割及提取中的应用研究_________

Application of graph theory in brain tumor segmentation and extraction

作者：李鹏吴水才高宏建盛磊

单位：           北京工业大学生命科学与生物工程学院(北京100124)    

关键词：图论；脑肿瘤；图像分割

分类号：

出版年·卷·期（页码）：2013·32·3（243-247）

摘要：

目的基于Matlab和VC++混合编程，实现了图论在脑肿瘤分割及提取中的应用，为之后脑肿瘤三维重建提供准确的分割结果。方法在Matlab和VC++开发平台下，首先读取含脑肿瘤的MRI图像，经过一定的预处理后，调用C++编写的图论分割函数，实现MRI图像的全局分割，然后通过肿瘤区域的颜色信息进行区域二值化和轮廓提取等后处理，很好地完成了脑肿瘤的分割提取。结果通过与专家手动分割的脑肿瘤区域进行比较以及对算法各模块运行时间的监测，显示脑肿瘤分割准确度高，且算法运行稳定。结论基于图论的分割算法能够反映图像全局特性，且运行稳定，是一种值得推广的脑肿瘤分割方法。

Objective Based on Matlab and VC++mixed programming，this paper realizes the application of graph theory in the brain tumor segmentation and extraction，providing accurate segmentation results for subsequent brain tumor three-dimensional reconstruction. Methods On Matlab and VC++development platform，the MR images with brain tumors are read firstly，after certain preprocessing，the graph theory segmentation functions written in C++are called to realize the global segmentation of MR images. Then some postprocessing including region binarization and contour extraction according to color information of tumor regions are done to complete the brain tumor segmentation and extraction. Results Compared with the manual segmentation of brain tumor region by expert，and with the monitoring on the running time of each module in the algorithm，the results are highly accurate in brain tumor segmentation and the segmentation algorithm runs stably. Conclusions The image segmentation algorithm based on graph theory reflects the global image properties，runs stably，and is worthy of popularization in brain tumor segmentation.

参考文献：

［1］黄峰茜. 脑部肿瘤图像分割技术的研究［D］. 南京：南京航空航天大学，2008.

Huang Fengqian. Research on division technology of brain tumor image［D］. Nanjing：Nanjing University of Aeronautics and Astronautics，2008.

［2］Felzenszwalb PE，Huttenlocher DP. Efficient Graph-Based Image Segmentation ［J］. International Journal of Computer Vision， 2004，59(2)：167-181.

［3］李素莹. 基于图论的乳腺肿瘤超声图像的分割和识别方法［D］. 广州：华南理工大学，2011.

Li Suying. A graph-based segmentation and recognition method for breast tumors in ultrasound images［D］. Guangzhou：South China University of Technology，2011.

［4］艾海明，吴水才，高宏建，等. 基于图论的肝肿瘤CT图像自动分割方法［J］. 北京工业大学学报， 2010，36 (4)：572-576.

Ai Haiming，Wu Shuicai，Gao Hongjian，et al. Graph-based method for liver tumor CT image auto-segmentation［J］. Journal of Beijing University of Technology，2010，36(4)：572-576.

［5］黄茜，杨文亮，顾杰峰. 一种改进的基于图论的图像分割方法［J］. 科学技术与工程，2009，9 (13)：3652-3656.

Huang Qian，Yang Wenliang，Gu Jiefeng. Improved image segmentation method based on graph theory［J］. Science Technology and Engineering，2009，9(13)：3652-3656.

［6］Deepthi Narayan，Srikanta MK，Kumar GH. Image Segmentation Based on Graph Theoretical Approach to Improve the Quality of Image Segmentation ［C］. World Academy of Science，Engineering and Technology， 2008，（42）：35-38.

［7］Ming Zhang，Reda Alhajj. Improving the Graph-Based Image Segmentation Method［C］. Proceedings of the 18th IEEE international Conference on Tools with Artificial Intelligence，2006：617-624.

［8］李伟，陈武凡. 基于图论的交互式脑部肿瘤MRI自动三维分割［J］. 南方医科大学学报，2009，29 (1)：140-143.

Li Wei，Chen Wufan. Graph-based interactive three-dimensional segmentation of magnetic resonance images of brain tumors ［J］. Journal South Medical University， 2009，29(1)：140-143.

［9］田丽霞. 基于图论的复杂脑网络分析［J］.北京生物医学工程，2010,29(1):96-100.

Tian Lixia. Analysis of complex brain networks based on graph theory［J］.Beijing Biomedical Engineering，2010,29(1):96-100.

［10］王海英. 图论算法及其MATLAB实现［M］. 北京：北京航空航天大学出版社，2010：1-2.

Wang Haiying. Graph theory algorithm and MATLAB realization［M］. Beijing：Beijng University of Aeronautics and Astronautics Press，2010：1-2.

［11］谭志明. 基于图论的图像分割及其嵌入式应用研究［D］. 上海：上海交通大学图像通信与信息处理研究所，2008.

Tan Zhiming. Research on graph theory based image segmentation and its embedded application［D］. Shanghai：Institute of Image Communication & Information Processing，Shanghai Jiao Tong University，2008.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录