摘要
医疗影像数据是医生进行临床诊断、跟踪病情发展、制定手术计划的重要客观依据.为了使影像数据的利用更加准确和高效,将纹理统计分析方法应用至胸主动脉CT图像的纹理特征分析中,以像素对间的方向参数和距离参数构造灰度共生矩阵,并从该矩阵中提取出有意义的统计量来表征纹理.实验证明,使用该共生矩阵提取出的特征能够区分正常主动脉和夹层主动脉,为进一步建立医学图像辅助诊断系统提供基础.
The massive patient image data is an important objective basis in clinical diagnosis, planning, tracking and operation research for doctors. In order to use the image data more efficient and correct, this article extracts the thoracic aorta CT image texture feature by statistical analysis. The symbiotic matrix is constructed by direction and distance parameter between pixels. Then matrix extracts from the energy, entropy, contrast, correlation and other meaningful statistics to represent the texture characteristics, Experiments proved that the feature extracted by the co-occurrence matrix can distinguish normal aorta and interlayer aorta, to provide the premise condition for further establish medical image aided diagnosis system.
作者
陈菁
林春深
CHEN Jing LIN Chunshen(School of Chemical Engineering, Fuzhou University, Fuzhou, Fujian 350116, China)
出处
《福州大学学报(自然科学版)》
CAS
北大核心
2017年第1期91-97,共7页
Journal of Fuzhou University(Natural Science Edition)
基金
国家质检总局科技计划资助项目(2010QK032)
关键词
CT图像
胸主动脉夹层
灰度共生矩阵
纹理特征
诊断技术
CT image
thoracic aortic dissection
gray-level co-occurrence matrix
textural feature
diagnostic technique