中国组织工程研究

中国组织工程研究 ›› 2010, Vol. 14 ›› Issue (35): 6544-6546.doi: 10.3969/j.issn.1673-8225.2010.35.021

• 骨与关节图像与影像 bone and joint imaging • 上一篇    下一篇

肝癌图像感兴趣目标的准确分割

史延新   

  1. 西安工业大学数理系，陕西省西安市  710032
  • 出版日期:2010-08-27 发布日期:2010-08-27
  • 作者简介:史延新☆，女，1968年生，山西省汾阳市人，汉族，1990年西安工业大学毕业，博士，高级工程师，主要从事模式识别、图像处理方面的研究。 shiyxx001@sina.com
  • 基金资助:

    高等学校博士学校点专项科研基金资助项目(20060699032)。

Segmentation of interesting targets in liver cancer image

Shi Yan-xin   

  1. Department of Mathematics and Physics, Xi’an Technology University, Xi’an  710032, Shaanxi Province, China
  • Online:2010-08-27 Published:2010-08-27
  • About author:Shi Yan-xin☆, Doctor, Senior engineer, Department of Mathematics and Physics, Xi’an Technology University, Xi’an 710032, Shaanxi Province, China shiyxx001@sina.com
  • Supported by:

    Doctoral Program Foundation of Institutions of Higher Education of China, No. 20060699032*

PDF

350

摘要：

背景：Snake模型为医学图像分割提供了一个全新的分割方式，可以克服传统图像分割方法在医学图像分割中的缺点。
目的：针对肝癌CT图像特点，提出了一种改进的B样条曲线的Snake模型图像分割算法。
方法：对腹部CT图像进行预处理，获得肝脏癌变部分的初始轮廓，再构造闭合B样条Snake模型，最后使用MMSE最小化外力变形模型以实现图像的准确分割。
结果与结论：改进的B-Snake分割算法不仅减少了噪声的影响，而且使Snake曲线较好地收敛于目标轮廓边缘，对于肝癌CT图像该方法取得了感兴趣目标的良好分割效果。

关键词: 肝癌, B样条曲线, CT, Snake模型, MMSE, 图像分割

Abstract:

BACKGROUND: Snake model is a new approach for medical image segmentation, which can overcome the defects of the traditional medical image segmentation.
OBJECTIVE: To propose an improved B-spline Snake model for image segmentation based on the features of liver cancer CT images.
METHEDS: The abdominal CT image was preprocessed to obtain initial outline of liver cancer parts, and then, closed B-spline Snake model was constructed. Finally, deformed model image of accurate segmentation was obtained using the MMSE to minisize extemal force.
RESULTS AND CONCLUSION: The improved B-Snake segmentation algorithm not only reduces noises, but also converge the Snake curve to target contour edge. It is an effective method for segmenting the interesting target of liver CT images.

中图分类号: