Machine Vision Laboratory, University of the West of England, Bristol BS16 1QY, UK; Faculty of Engineering, University of Leeds, Leeds LS2 9JT, UK.
Comput Med Imaging Graph. 2011 Mar;35(2):155-65. doi: 10.1016/j.compmedimag.2010.10.004. Epub 2010 Nov 11.
This article describes an enhanced curvature pattern based melanoma diagnosis system using convolution techniques and ensemble classifiers. We extract the 3D data of melanoma with a photometric stereo device first. Then differential forms of the melanoma surface can be extracted with the convolution method proposed. After extracting 3D based differential forms, statistical moments of enhanced principal curvatures of skin surfaces are calculated to describe the geometrical texture patterns. Finally, ensemble classifiers are constructed whose optimal mean sensitivity and specificity can reach 89.24 percent and 87.62 percent respectively. Comparisons with skin tilt/slant pattern based 3D shape characterization method and 2D methods like color variation and border irregularity are also included.
本文提出了一种基于增强曲率模式的黑色素瘤诊断系统，该系统使用卷积技术和集成分类器。我们首先使用光度立体设备获取黑色素瘤的 3D 数据。然后，利用所提出的卷积方法提取黑色素瘤表面的微分形式。在提取基于 3D 的微分形式后，计算皮肤表面增强主曲率的统计矩以描述几何纹理模式。最后，构建集成分类器，其最佳平均灵敏度和特异性分别可达 89.24%和 87.62%。还与基于皮肤倾斜/斜率模式的 3D 形状特征化方法以及 2D 方法（如颜色变化和边界不规则性）进行了比较。
