基于数据网格的肝脏CT图像感兴趣区域分割

通过图像分割获取医学图像感兴趣区域是医学图像处理与分析要解决的首要问题及技术难点。为了获取肝脏CT图像中的感兴趣区域,针对肝脏CT图像的自身特点,提出了一种基于数据网格和改进的区域生长法相结合的肝脏CT图像感兴趣区域分割方法。首先利用数据网格对肝脏CT图像做粗分割,最大程度地去除骨骼及背景对后期分割的影响。其次,利用先验知识,确定区域生长的种子点,做基于改进的区域生长法的二次分割。实验结果表明,能较完整的分割出CT图像的肝脏区域。实验结果为进一步的医学肝脏图像的处理与分析奠定了基础。

ROI Segmentation of Liver CT Images Based on Data Grid

The ROI (Region of Interest) of medical images can be got through image segmentation, and it is the first problem and the key technology to be solved for medical image processing and analysis. In order to get the ROI of liver CT images, an ROI segmentation method is proposed according to the characteristics of liver CT images. It is based on the combination of data grid and region growing algorithm. Firstly, the data grid method is used to roughly segment the liver CT images which can furthest removal the effect of skeletal and background on later segmentation. Secondly, the seed of region growing can be determined using the prior knowledge, and the second segmentation can be done based on region growing. The experiment results show that the method can completely segment the ROI of liver CT images and lay a foundation for further image processing and analysis.