利用多分辨率直方图特征分类数字X光乳腺图像

刘欣悦; 黄廉卿

doi:null

您当前的位置：

首页 >

文章列表页 >

利用多分辨率直方图特征分类数字X光乳腺图像

更新时间：2020-08-12

- 利用多分辨率直方图特征分类数字X光乳腺图像
- Classification of digital mammograms using multi-resolution histogram features
- 光学精密工程 2006年14卷第2期页码：327-332
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所,吉林长春,中国,130033
  2. 中国科学院研究生院北京,100039
- 作者简介：
- 基金信息：
- DOI：
  中图分类号： TP391.4
- 收稿日期：2005-11-14，
  
  修回日期：2006-01-22，
  
  网络出版日期：2006-04-30，
  
  纸质出版日期：2006-04-30
- 稿件说明：
移动端阅览
刘欣悦, 黄廉卿. 利用多分辨率直方图特征分类数字X光乳腺图像[J]. 光学精密工程, 2006,14(2):327-332.

LIU Xin-yue, HUANG Lian-qing. Classification of digital mammograms using multi-resolution histogram features[J]. Optics and precision engineering, 2006, 14(2): 327-332.
刘欣悦, 黄廉卿. 利用多分辨率直方图特征分类数字X光乳腺图像[J]. 光学精密工程, 2006,14(2):327-332. DOI：

LIU Xin-yue, HUANG Lian-qing. Classification of digital mammograms using multi-resolution histogram features[J]. Optics and precision engineering, 2006, 14(2): 327-332. DOI：

摘要

提出了一种结合多分辨率直方图特征表示与核学习算法的数字X光乳腺图像的分类方法。该方法不依赖特征选择步骤

而是基于感兴趣区(ROI)的高维多分辨率直方图特征

通过从训练实例中学习

同时检测多种异常的ROI。对该方法进行接收器工作特性(ROC)分析

敏感性约为89%

ROC曲线下面积(AUC)接近0.91。与以前所提出的检测方法相比

该方法不需要针对特定类型病变选择特征表示

因此可以同时检测多种类型的病变

简化了检测过程

提高了检测效率

而且分类性能也达到或超过了以前方法的平均分类性能。结果表明

利用多分辨率直方图特征表示能够很好地区分乳腺图像中正常和异常区域

同时也显示了借助核学习算法消除或限制分类任务中特征选择步骤的可能性。

Abstract

A classification approach of digital mammograms using multi- resolution histogram representation in conjunction with kernel-based learning methods was presented. The approach didn't rely on the feature selection step and learned to classify various kinds of Region Of Interest (ROI) as normal/abnormal using its high-dimensional multi-resolution histogram features. Receiver Operating Characteristic (ROC) analysis of classification performance of the proposed approach shows that the sensitivity is about 89% and the Area Under Curve (AUC) is nearly 0.91. Compared to previous approaches

the proposed approach does not need to select abnormality-specific features so that it can detect various kinds of abnormalities simultaneously

which simplifies the detection process and improves the detection efficiency. The results demonstrate that multi-resolution histogram features can clearly distinguish the normal or abnormal classes in mammograms and the feature selection step of certain classification tasks can be eliminated or limited by using kernel-based learning method.

关键词

Keywords

references

. KARSSEMEIJER N,HENDRIKS J H.Computer- assisted reading of mammograms[J]. Europe Radiology,1997,(7):743-48.

. JIANG Y, NISHIKAWA R M,SCHMIDT RA, et al.Improving breast cancer diagnosis with computer-aided diagnosis[J].Academic Radiology,1999,6, 22-33.

. Breast Imaging Reporting and Data System, Third Edition, American College of Radiology[Z]. 1998.

. BAKER J A,KORNGUTH P J,LO J Y,et al.Breast cancer: prediction with artificial neural network based on BI-RADS standardized lexicon[J].Radiology,1995,196.

. SONKA M,FITZPATRICK J M. Computer-aided diagnosis in mammography . Handbook of Medical Imaging. SPIE Press, 2000.

. El-NAQA I, YANG Y, WERNICK M N,et al.A support vector machine approach for detection of microcalcifications[J].IEEE Trans. Medical Imaging,2002,21:1552-63.

. CHANG R F, WU W J, MOON W K,et al.Support vector machines for diagnosis of breast tumors on US images[J].Academic Radiology, 2003,16:189-197.

. WEI L Y,YANG Y Y,NISHIKAWA R M,et al.A Study on several machine-learning methods for classification of malignant and benign clustered microcalcifications[J].IEEE Trans. Medical Imaging, 2005,24:371-80.

. CHAPELLE O,HAFFNER P,VAPNIK V.SVMs for histogram-based image classification[J]. IEEE Trans. Neural Networks,1999,10:1055-1065.

. HADJIDEMETRIOU E,GROSSBERG M D,NAYAR S K.Multiresolution histogram and their use for recognition[J].IEEE Trans. Pattern Analysis and Machine Intelligence,2004,26:831-74.

. SPORRING J,WEICKERT.Information measures in scale-spaces[J].IEEE Trans. Information Theory,1999,45:1051-58.

. VAPNIK V. Statistical learning theory[M]. New York: Wiley, 1998.

. MANGASARIAN O L.Generalized support vector machines[M]. Advances in Large Margin Classifiers, MIT Press, 2000.

. TIPPING M E.Sparse Bayesian learning and the relevance vector machine[J]. J. Machine Learning Research,2001,(1):211-44.

. SUCKLING J,PARKER J,DANCE D R,et al.The mammographic image analysis society digital mammogram database .2nd International Workshop on Digital Mammography, 375-78, York, England, Elsevier, 1994.

. MULLER K R,MIKA S,RATSCH G.An introduction to kernel-based learning algorithms[J].IEEE Trans. Neural Networks, ２００１,12:181-201.

. PEPE M S.Receiver operating characteristic methodology[J].J. American Statistical Association,2000,95:308-311.

. FAWCETT T.ROC graphs:notes and practical considerations for researchers[J]. Technical Report, HP Laboratories, Palo Alto, CA, USA, April 2004.

. CHU Y,LI L H,GOLDGOF D,et al.Classification of masses on mammograms using support vector machine[J].SPIE,2003,5032:940-948.

. SUN X J, QIANG W, SONG D S.Three-class classification in computer-aided diagnosis of breast cancer by support vector machine[J].SPIE,2004,5370:999-1007.

浏览量

436

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于监督保局子空间虚假近邻准则的原始特征选择

基于边界的最大间隔模糊分类器

用于分类的样本保局鉴别分析方法

胸部CT图像中孤立性肺结节良恶性快速分类