MR脑肿瘤图像分割与矩特征研究  

MR brain noeplasms image division and moment characteristic research

作　　者：俞海平[1] 董育宁[2] 邬立保[2] 丁智[2] 李茗[1] 

机构地区：[1]南京大学医学院附属鼓楼医院放射科,江苏南京210008 [2]南京邮电大学通信与信息工程学院,江苏南京210003

出　　处：《中国医学影像技术》2008年第8期1274-1276,共3页Chinese Journal of Medical Imaging Technology

基　　金：南京社会发展基金资助项目(ykk05108)

摘　　要：目的对MR脑肿瘤图像进行分割,并对分割的结果进行矩描述。方法在分析当前常用的医学图像分割方法的基础上,提出一种基于形变模型的医学图像分割方法,并给出了相应的理论算法模型和实现步骤,最后用Visual C++6.0编程,并对MR脑肿瘤图像进行分割实验。结果本文分割方法分割边界清晰,总体不确定性较小。结论本文分割方法切实可行,分割效果较好,为进一步的MR脑肿瘤图像分析和研究提供了一种有效工具。Objective To segmentate MR brain noeplasms image,and to moment describe segmentation of the results.Methods After analyzing the medical image segmentation methods that are widely used now,segmentation method based on deformation model of medical images was presented,and the relevant theoretic arithmetic model and realization protocol were summed up.Using the experiment data for MR brain noeplasms image,the imaging segment code was carried out with Visual C＋＋6.0.Results The result of this method is very distinct,and whole uncertainty is relative low.Conclusion The segmentation technology mentioned in this paper is feasible,and the effect of segmentation is preferable,so it provides an efficient tool to analyze MR brain noeplasms images and investigate disease farther.

关 键 词：SNAKE模型 梯度矢量流 能量函数 特征提取 LEGENDRE矩 

分 类 号：R445.2[医药卫生—影像医学与核医学]