基于图像跟踪的呼吸运动分析

呼吸运动是一种内在的、周期性运动,传统呼吸运动模型用公式描述,需要估计人的运动周期和运动范围,无法满足三维和实时分析的需要。本文提出一种新的呼吸运动分析方法,该方法借助于用X光机实时采集的患者胸部透视序列图像,通过一种改进的块匹配方法跟踪图像中目标点(横膈膜)的位置变化建立呼吸运动模型。实验证明:与传统的呼吸运动模型相比,该方法更能实时反映人体呼吸运动规律,并且方便建立三维呼吸运动模型,具有很高的临床实用价值。

Breath motion estimation based on image tracking

The organ motion due to respiration is an intra treatment motion,the traditional breath motion can be described by mathematical model in some special cases with only the fixed value of the extent of the motion and the period of breathing cycle,which can't build 3D model and for online analysis.In this paper we proposed a new model by tracking the position of the interest point of the image set.The experiment proves that it can reflect the vary of motion and easy to build 3D breath motion model.