电子学报 ›› 2018, Vol. 46 ›› Issue (6): 1319-1326.DOI: 10.3969/j.issn.0372-2112.2018.06.007

• 学术论文 • 上一篇    下一篇

一种全自动的肺裂分割方法

彭圆圆, 肖昌炎   

  1. 湖南大学电气与信息工程学院, 湖南长沙 410006
  • 收稿日期:2016-07-05 修回日期:2017-10-16 出版日期:2018-06-25
    • 通讯作者:
    • 肖昌炎
    • 作者简介:
    • 彭圆圆,男,1987年出生,湖北武穴人,博士研究生,主要研究方向为医学图像处理、模式识别.
    • 基金资助:
    • 国家自然科学基金 (No.61172160,No.61571184)

Automatic Segmentation of Pulmonary Fissures in CT Chest Images

PENG Yuan-yuan, XIAO Chang-yan   

  1. College of Electrical and Information Engineering, Hunan University, Changsha, Hunan 410006, China
  • Received:2016-07-05 Revised:2017-10-16 Online:2018-06-25 Published:2018-06-25
    • Corresponding author:
    • XIAO Chang-yan
    • Supported by:
    • National Natural Science Foundation of China (No.61172160, No.61571184)

摘要: CT(Computer Tomography)图像中自动分割肺裂是很困难的,肺裂往往存在不完整、形变、断裂和附裂等现象.本文提出一种融合肺部解剖结构特征来实现自动分割肺裂的方法.首先结合肺部气管和动脉血管信息定位肺裂感兴趣区域.然后利用肺裂方向信息增强肺裂,并利用多剖面滤波器滤除噪声从而对肺裂进行预分割.最后融合已定位的肺裂感兴趣区域和肺裂预分割结果来自动分割肺裂.与人工参考对比,提出的算法在人体左肺和右肺中分割的肺裂的F1-score中值分别为0.881和0.878.

关键词: 自动分割肺裂, 肺裂感兴趣区域, 肺裂预分割, 多剖面滤波器

Abstract: Automatic segmentation of pulmonary fissures is a nontrivial task in CT(Computer Tomography) chest images,due to incomplete,disrupted,deformed and accessory fissures.In this paper,we present a approach to fuse pulmonary structure characteristics for fissure segmentation.Firstly,we fuse the prior knowledge of trachea and pulmonary arteries to identify fissure region of interest.Then fissures directional field is exploited to enhance fissures and a multi-plane filter is proposed to remove noise for fissure pre-segmentation.Finally fissure region of interest and fissure pre-segmentation are combined for fissure segmentation.Compared with manual fissure references,our method obtained a high segmentation accuracy with median F1-score of 0.881 and 0.878 for the left and right lung images respectively.

Key words: automatic segmentation of pulmonary fissures, fissure region of interest, fissure pre-segmentation, multi-plane filter

中图分类号: