基于迭代算法的CT肺部兴趣区图像提取

Extraction of Region of Interest from Lung CT Images Based on Iterative Algorithm

目的自动提取CT肺部兴趣区,以辅助医生诊断和治疗肺部疾病。方法利用迭代算法自动提取肺部兴趣区,并和影像医生手工提取的肺部兴趣区作对比。结果自动提取的肺部兴趣区图像与手工提取的肺部兴趣区图像在外形、灰度和方差方面,结果非常接近。结论利用迭代算法自动提取肺部兴趣区的方法可行。

Objective Extracting region of interest（ROI） from lung CT images automatically to assist doctors to diagnose and treat lung diseases. Methods Iterative algorithm is used to extract lung ROI automatically, which was then compared with the ROI extracted manually by a radiologist. Results Lung ROI extracted by the two different methods are extremely close in the aspects of shape, grayscale and variance. Conclusion Using iterative algorithm to extract lung ROI is feasible.