期刊
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
卷 8, 期 6, 页码 688-696出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.bspc.2013.06.011
关键词
Shearlet transform; Texture feature; Breast tumor; Ultrasound image; Classification
资金
- Innovation Program of Shanghai Municipal Education Commission [09YZ15]
- National Natural Science Foundation of China [60701021]
- Leading Academic Discipline Project of Shanghai Educational Committee [J50104]
To augment the classification accuracy of the ultrasound computer-aided diagnosis (CAD) for breast tumor detection based on texture feature, we proposed to extract texture feature descriptors by the shearlet transform. Shearlet transform provides a sparse representation of high dimensional data with especially superior directional sensitivity at various scales. Therefore, shearlet-based texture feature descriptors can characterize breast tumors well. In order to objectively evaluate the performance of shearlet-based features, curvelet, contourlet, wavelet and gray level co-occurrence matrix based texture feature descriptors are also extracted for comparison. All these features were then fed to two different classifiers, support machine vector (SVM) and AdaBoost, to evaluate the consistency. The experimental results of breast tumor classification showed that the classification accuracy, sensitivity, specificity, positive predictive value, negative predictive value and Matthew's correlation coefficient of shearlet-based method were 91.0 +/- 3.8%, 92.5 +/- 6.6%, 90.0 +/- 3.8%, 90.3 +/- 3.8%, 92.6 +/- 6.3%, 0.822 +/- 0.078 by SVM, and 90.0 +/- 2.8%, 90.0 +/- 4.0%, 90.0 +/- 23%, 89.9 +/- 2.4%, 90.1 +/- 3.6%, 0.803 +/- 0.056 by AdaBoost, respectively. Most of the shearlet-based results significantly outperformed those of other method based results under both the classifiers. The results suggest that the proposed method can well characterize the properties of breast tumor in ultrasound images, and has the potential to be used for breast CAD in ultrasound image. (C) 2013 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据