摘要
目的:在利用光学相干断层成像技术检测和诊断皮肤癌时,获取能够用于皮肤癌诊断的图像特征值。方法:应用差分盒子技术方法,提取色素痣、基底细胞癌和黑色素瘤的光学相干断层图像的二维分形维数。结果:从光学相干断层图像中提取出的二维分形维数能够有效地区分色素痣、基底细胞癌及黑色素瘤,色素痣和基底细胞癌的分形维数较黑色素痣小,具有统计学意义。结论:应用本方法计算的二维分形维数可以用于皮肤癌的诊断。
Objective To obtain quantitative imaging features for diagnosing the skin tumors with optical coherence tomography（OCT）. Methods The differentiate box counting method was use to calculate the two-dimensional fractal dimension from OCT images of the pigment nevi, basal cell carcinomas and melanomas. Results The pigment nevi, basal cell carcinomas and melanomas were effectively classified using the two-dimensional fractal dimension calculated from OCT images. The fractal dimension of pigment nevi and basal cell carcinomas was smaller than that of melanomas, with statistical differences. Conclusion The two-dimensional fractal dimension calculated from OCT images can be used in the diagnosis of skin tumors.
出处
《中国医学物理学杂志》
CSCD
2017年第7期698-701,共4页
Chinese Journal of Medical Physics
基金
宁波工程学院科研项目(D2016009)
宁波工程学院大学生科技创新项目(应用光学相干断层成像技术量化检测皮肤癌)
关键词
光学相干断层成像
分形维数
差分盒子计数法
皮肤癌
图像处理
optical coherence tomography
fractal dimension
differentiate box counting method
skin tumor
image processing
