摘要
肺结节的三维可视化及其体积大小的变化,有利于医学研究和临床诊断。提出一种改进的基于蒙特卡洛的肺结节体积测量算法,首先对肺部二维图像进行预处理,然后区域聚类分割得到肺实质并三维重建,接着使用三维区域增长法分割得到肺结节,最后采用改进的蒙特卡洛方法测量肺结节的体积。初步的实验结果表明,提出的算法与传统的蒙特卡洛(MC)和拟蒙特卡洛(QMC)算法相比较,测量得到的肺结节体积更加接近真实值,且相对误差最小,表明本文算法的优越性。
Three dimensional visualization cial to clinical diagnosis and medical research. Pulmonary nodule volume algorithm based on ted,segment lung parenchyma and 3D surface and size change of pulmonary nodules, are benefi- Therefore, this article puts forward an improved Monte Carlo. First, the image slices are pretrea- rendering reconstruction are performed. Next, the 3D region growing method is used to segment the pulmonary nodules. Finally, the improved Monte Carlo（MC） is used to measure the volume. The experimental results show that compared with the MC and Quasi-Monte Carlo （QMC） methods, the algorithm proposed in this paper is closer to the real value in the volume measurement of lung nodules, the relative error is smaller, therefore, the superiority of the algorithm is proved.
作者
周翔飞
强彦
ZHOU Xiangfei QIANG Yan(College of Computer Science and Technology, Taiyuan University of Technology, Taiyuan 030024,Chin)
出处
《太原理工大学学报》
CAS
北大核心
2016年第1期75-79,共5页
Journal of Taiyuan University of Technology
基金
国家自然科学基金资助项目:基于混合成像的孤立性肺结节计算机辅助诊断方法(61202163)
基于医学影像结构和功能混合特征的周围型肺癌计算机辅助诊断方法(61373100)
虚拟现实技术与系统国家重点实验室基金资助项目(BUAA-VR-15KF02)
关键词
肺结节
三维可视化
体积测量
蒙特卡洛
pulmonary nodules
three-dimensional visualization
volume measurement
Monte Carlo
