华中科技大学学报：自然科学版2011，Vol.39Issue(7)：26-29,4.
CT图像中肺结节毛刺特征的计算机量化方法
Computerized quantification method for lung nodule spiculation features on CT images
摘要
Abstract
An approach to quantify the lung nodule speculation levels on CT （computed tomography） images was proposed. The lung nodule was segmented by using the dynamic programming algorithm on the region of the interest. Based on the segmented boundary and the regularity of the gradients, a new index describing the perpendicularity between the normal vectors of the boundary and the gradients nearby the boundary was extracted to quantify the spiculation levels of the lung nodule. The experiment results indicate that this index can evaluate the spiculation levels accurately with a relatively high resolution to classify the non-speculation, short-speculation, and long-speculation lung nodules.关键词
计算机量化/肺结节/CT图像/毛刺特征/分割/边界法线/梯度Key words
computerized quantification/lung nodules/CT （computed tomography）/spiculation fea-tures~ segmentation/normal vectors of boundary~ gradient分类
信息技术与安全科学引用本文复制引用
王倩,龚建平,宋恩民,熊伟..CT图像中肺结节毛刺特征的计算机量化方法[J].华中科技大学学报：自然科学版,2011,39(7):26-29,4.基金项目
苏州市科技发展计划（国际科技合作）资助项目 （）
中南财经政法大学引进人才振兴工程科研启动基金资助项目（31140911305）. （）
