基于视觉感知信息的乳腺钼靶肿块检测分析与自动提取

doi:10.3969/j.issn.0258-8021. 2014.01.005

摘要
图/表
参考文献
相关文章 (5)

全文: PDF (2010 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要医生的视觉感知信息与影像诊断联系紧密，如何有效利用该信息以提高医学影像辅助诊断中的决策准确性，是一个具有前沿性和实际临床价值的研究课题。针对医生临床诊断时其视觉感知行为的分析和利用，探讨医生读片时单纯的视觉注视信息在多大程度上反映肿块位置（可检测性），以及如何利用视觉注视信息提取病灶。首先，用眼动仪采集医生读片时的注视点序列，每个注视点包括该点在钼靶影像中的相对位置、注视点停留时间和瞳孔直径等3个视觉特征，然后基于这些特征对注视点序列进行聚类分析，根据关注度评价找出医生浏览影像时的若干“关注点”位置，并对比分析其与肿块位置的关系，以评价“命中率”；以关注点为引导，利用区域生长和水平集方法对肿块病灶进行提取。利用DDSM数据库和浙江省肿瘤医院数据库的75张钼靶影像进行初步实验。将关注点数限制在4个以内时，肿块病灶命中率为58.49%，同时所有命中肿块中被完整提取的占70.97%。结果表明，医生视觉注意信息对肿块位置的反应有一定作用，有助于理解感知反馈提高诊断精度的内在机理。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	柯尔挺¹厉力华^1*刘伟¹徐伟栋¹

关键词 ：视觉感知, 图像分割, 眼动仪, 聚类分析

Abstract：Clinical diagnosis according to medical images is a process of radiologists’ visual perception and decisionmaking. The radiologists’ visual perception information is intimately associated with diagnosis. How to effectively use visual perception information to improve the decisionmaking accuracy in computeraided diagnosis is a research subject which is full of scientific significance and clinical value. This paper conducted researches on the analysis and the use of visual perception behavior during diagnosis to explore two issues:one is how well the single perceptual information during diagnosis can reflect masses position; the other is how to use the perceptual information for extracting masses. In the paper, the research method includes two steps. Firstly, radiologists’ fixation point sequence, in which every point includes fixation point location in mammogram, duration time and pupil diameters, was recorded by an eyetracker during reading and then clustered to achieve some radiologists’ concerns according to the three visual features. Shooting average was calculated by analyzing the positional relationship between concerns and masses in the same mammogram. Secondly, regarding concerns as seeds, the SBRG (seedsbased region growing) approach and the multiscale mass segmentation approach were applied to buckle breast masses from mammograms. The result of applying the proposed method to 75 mammograms from both DDSM and Zhejiang Cancer Hospital showed that it could achieve shooting average of 58.49% when the limitation of concerns number was 4, and the fullextraction rate for the shot masses was 70.97%. It is revealed that the perceptual information is helpful to reflect masses position as well as to understand the inner mechanism of perceptual feedback.

Key words： CAD visual perception image segmentation eyetracker clustering analysis

基金资助:国家重点基础研究发展计划资助(2013CB329502)；国家自然科学基金（61001215、61271063）

引用本文:

柯尔挺¹厉力华^1*刘伟¹徐伟栋^{1 . 基于视觉感知信息的乳腺钼靶肿块检测分析与自动提取[J]. 中国生物医学工程学报, 2014, 33(1): 28-36.}

链接本文:

http://cjbme.csbme.org/CN/ 10.3969/j.issn.0258-8021. 2014.01.005 或 http://cjbme.csbme.org/CN/Y2014/V33/I1/28

［1］A Jemal, R Siegel, E Ward, et al. Cancer statistics, 2008 ［J］. Caa Cancer Journal for Clinicians, 2008, 58(2): 71-96.
［2］报告称癌症成北京人主要死因，乳腺癌及肺癌居首［EB/OL］. http://news.sohu.com/20111011/n321857525.shtml, 2011-10-11/2012-07-01.
［3］谢凯涛. 乳腺癌和乳腺病［M］. 北京：北京医科大学中国协和医科大学联合出版社, 1995.
［4］刘佩芳. 乳腺影像诊断必读［M］. 北京：人民军医出版社, 2007.
［5］Al Sharkawy M, Sharkas M, Ragab D. Breast cancer detection using support vector machine technique applied on extracted electromagnetic waves ［J］. Applied Computational Electromagnetics Society Journal, 2012, 27(4): 292-301.

［6］Xu Xiangmin, Zhou Fengle, Xiao Yue. The study of mammary disease CAD system ［C］ // 2007 IEEE/ICME International Conference on Complex Medical Engineering. Beijing:IEEE,2007: 676-679.
［7］Mello-Thoms C, Dunn SM, Nodine CF, et al. The perception of breast cancersa spatial frequency analysis of what differentiates missed from reported cancers ［J］. IEEE Transactions on Medical Imaging, 2003, 22(10): 1297-1306.
［8］Kundel HL, Nodine CF, Krupinski EA, et al. Using gazetracking data and mixture distribution analysis to support a holistic model for the detection of cancers on mammograms ［J］. Academic Radiology, 2008, 7(15): 881-886.
［9］Kundel HL, Nodine CF, Conant EF, et al. Holistic component of image perception in mammogram interpretation: gazetracking study ［J］. Radiology, 2007, 242(2): 396-402.
［10］Krupinski EA. The importance of perception research in medical imaging ［J］. Radiation Medicine, 2000, 18(6): 329-334.
［11］Hu Kai, Gao Xieping, Li Fei. Detection of suspicious lesions by adaptive thresholding based on multiresolution analysis in mammograms ［J］. IEEE Transactions on Instrumention and Measurement, 2011, 60(2): 462-472.
［12］徐伟栋, 刘伟, 厉力华, 等. 一种基于海岛冲刷模型的乳腺钼靶肿块检测新算法［J］. 中国生物医学工程学报, 2011, 30(4): 567-575.
［13］Marcin Bator MN. Detection of cancerous masses in mammograms by template matching: Optimization of template brightness distribution by means of evolutionary algorithm ［J］. Journal of Digital Imaging, 2012, 25(1): 162-172.
［14］Robert MN, Lorenzo LP. Fundamental limitations in developing computeraided detection for mammography ［J］. Nuclear Instruments and Methods in Physics Research, 2011, 648(Supp 1): S251-S254.

［15］虞红伟, 厉力华, 徐伟栋, 等. 一种基于水平集的多尺度乳腺肿块分割方法［J］. 仪器仪表学报, 2010, 31(6): 1418-1423. 
［16］Applied Science Laboratories ［EB/OL］. http://asleyetracking.com, 2011-03-01/2012-07-01. 
［17］Ester M, Kriegel H P, Sander J, et al. A densitybased algorithm for discovering clusters in large spatial data sets with noise ［C］// International Conference on Knowledge Discovery and Data Mining. Oregon:KDD, 1996: 226-231.
［18］夏鲁宁, 荆继武. SADBSCAN：一种自适应基于密度聚类算法［J］. 中国科学院研究生院学报, 2009, 26(4): 530-538.
［19］李双, 刘伟, 徐伟栋, 等. 基于眼动感知信息的彩色图像感兴趣区域提取算法［J］. 光电工程, 2011, 38(4): 115-123.