摘要： 借助手术导航技术的微创消融将逐渐成为外科治疗房颤的主流方法。快速、准确地分割心室、心房是微创消融的关键技术之一。针对心脏CT图像的具体特征提出一种分割技术，将基于像素的传统方法和基于水平集的活动轮廓模型相结合，完成图像预分割。分割结果直接作为变分水平集的初始轮廓，经若干次迭代后，得到心室、心房的光滑轮廓。

关键词: 微创消融, 图像分割, 水平集, 初始轮廓

Abstract: Minimally invasive ablation using surgical navigation technology gradually becomes the main-stream method in atrial fibrillation surgery. Extracting ventricle and atria out fastly and accurately is key technology in minimally invasive ablation. According to heart CT image’s characteristics, a segmentation method, which combines traditional pixel-based method with level-set-based active contour model is proposed to do pre-segmentation. The segmentation result is directly used as the initial contour of the variational level set. The curve smoothly matches the edge of object after several iterations.

Key words: minimally invasive ablation, image segmentation, level set, initial contour

中图分类号: 

• TP391.41