基于新过渡区提取法的医学图像交互式分割

目的：为了提高医学图像三维重组过程中二维图像交互分割的效率,我们试验了一种新过渡区提取方法,并验证了其分割效率和分割效果。方法：新过渡区提取法先依据原始图像,分散构造过渡片段,后将过渡片段按一定规则连接成过渡区,在过渡区的基础上实现交互分割。以512×512的CT图像序列为样本进行实验,结果与基于边缘提取的交互分割法及完全手工分割法比较。结果：基于新过渡区提取的交互分割法对目标区域的分割精确度高于边缘提取分割法,更接近完全手工分割,而人工操作时间较基于边缘提取的分割法及完全手工分割法分别节省39%和62%。结论：新过渡区提取法保留了过渡区分割法抗干扰能力强、对弱边界提取效果好的优点。应用于交互分割,能够减少人工劳动,提高分割效率,同时分割精确度高于边缘提取法,重组后能更准确反映目标特征。

A new method of interactive segmentation based on transition region extraction

Objective：In order to improve the efficiency of interactive segmentation in 3D reconstruction of medical images, we developed a new method of transition region extraction and verify its effect and efficiency. Methods：First, transition segments were constructed based on original image, then they were connected up to build transition regions. A serial CT images of 512 X 512 were employed to test how this new method functioned as the basis of interactive segmentation comparing with the interactive segmentation based on edge detection and manual segmentation. Results：Interactive segmentation based on this new method segment images was more accurately than that based on edge detection, and saved about 39% and 62% manual operation time comparing with the other two methods respectively. Conclusion： Such a method in- herits the character of transition region extraction. The application of interactive segmentation is helpful in reducing labor and improving efficiency of segmentation, meanwhile, it improves, segmentation accuracy. The effects of 313 reconstruction show the characteristics of objects more precisely.