Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (9): 142-145.DOI: 10.3778/j.issn.1002-8331.2010.09.040

• 图形、图像、模式识别 • Previous Articles     Next Articles

Liver MR image segmentation with iterative quadtree decomposition

CHI Dong-xiang1,CHENG Wei-zhong2   

  1. 1.School of Electronics & Information,Shanghai Dianji University,Shanghai 200240,China
    2.Department of Radiology,Affiliated Zhongshan Hospital,Fudan University,Shanghai 200032,China
  • Received:2009-12-15 Revised:2010-01-30 Online:2010-03-21 Published:2010-03-21
  • Contact: CHI Dong-xiang



  1. 1.上海电机学院 电子信息学院,上海 200240
    2.复旦大学 附属中山医院 放射科,上海 200032
  • 通讯作者: 迟冬祥

Abstract: Magnetic Resonance(MR) images are generally acknowledged as a golden standard for diagnosis of liver tumors.Therefore a good liver segmentation result is very important to computer-aided-diagnosis.Because of infiltrations among the internal organs and their individual differences,it is difficult for a liver segmentation and there is no universal medical method of segmentation.In this paper,on the basis of formal research,an iterative quadtree decomposition algorithm is put forward and liver image is segmented automatically.Experimental results show its feasibility and advantages and lay a foundation for the tumor extraction.

Key words: Magnetic Resonance Image(MRI), iterative quadtree decomposition, liver segmentation

摘要: 磁共振(Magnetic Resonance,MR)图像的诊断是公认的确认肝脏有无肿瘤等器质性病变的金标准方法,因此肝脏的正确分割对计算机辅助诊断有非常重要的意义。由于脏器组织浸润和个体差异,在肝脏分割实现方法方面有一定难度,目前尚没有通用的医学分割方法。在既有研究的基础上,提出了基于四叉树的迭代分割算法,得到MR图像中肝脏的自动分割结果。实验分割结果表明这种方法的可行性和优势,并为后续的肿瘤提取奠定基础。

关键词: 磁共振图像, 迭代四叉树, 肝脏分割

CLC Number: