A methodological approach to the classification of dermoscopy images
A methodological approach to the classification of dermoscopy images
复制标题
DOI：
10.1016/j.compmedimag.2007.01.003
复制
发表时间：
2007-09-01
影响因子：
5.7
通讯作者：
Moss, Randy H.
中科院分区：
文献类型：
--
作者：
Celebi, M. Emre;Kingravi, Hassan A.;Moss, Randy H.
本文提出了一种对皮肤镜图像中色素性皮肤病变进行分类的方法。首先，进行自动边界检测以将病变从背景皮肤中分离出来。然后从该边界提取形状特征。为了提取与颜色和纹理相关的特征，使用欧几里得距离变换将图像划分为多个具有临床意义的区域。将这些特征数据输入到一个优化框架中，该框架使用各种特征选择算法对特征进行排序，并根据从支持向量机分类获得的受试者工作特征曲线下面积度量来确定最佳特征子集大小。使用各种采样策略解决类别不平衡问题，并使用蒙特卡罗交叉验证估计分类器的泛化误差。对564张图像进行的实验得出特异性为92.34%，敏感性为93.33%。（c）2007爱思唯尔有限公司。保留所有权利。
In this paper a methodological approach to the classification of pigmented skin lesions in dermoscopy images is presented. First, automatic border detection is performed to separate the lesion from the background skin. Shape features are then extracted from this border. For the extraction of color and texture related features, the image is divided into various clinically significant regions using the Euclidean distance transform. This feature data is fed into an optimization framework, which ranks the features using various feature selection algorithms and determines the optimal feature subset size according to the area under the ROC curve measure obtained from support vector machine classification. The issue of class imbalance is addressed using various sampling strategies, and the classifier generalization error is estimated using Monte Carlo cross validation. Experiments on a set of 564 images yielded a specificity of 92.34% and a sensitivity of 93.33%. (c) 2007 Elsevier Ltd. All rights reserved.