基于双Snake模型的颅骨CT图像分割方法及应用

摘要

A method which using double-snake model segmentation algorithm to determine internal and external borders of the skull in the CT images was proposed,using polygon to approach its borders ,then by calculating the distance between the line segments of inner and outer polygons to get the data of the thickness of pre-frontal ,parietal and occipital bones in image.So it can provide a good method to the application of getting thickness information of skull base on CT image. Following an analysis of the thickness data of pre-frontal ,parietal and occipital bones at different positions ,the standard deviation and the rate of change of the thickness of parietal and occipital bones were proved to be larger. The proposed method can describe the boundary of the skull accurately,this method will provide a basis to establish a database of human heads and research the biomechanics modeling of head and so on.%利用活体人颅骨CT图像,通过双snake模型边界分割算法确定CT图像中颅骨的内外边界,并利用多边形近似其边界,随后通过计算内外多边形中相对应的线段间的距离,实现前额骨、顶骨和枕骨在图像中的厚度信息的获取,从而为基于CT图像获取颅骨的厚度信息的应用提供了一种有效手段.在对前额骨、顶骨和枕骨的不同位置的厚度计算结果进行分析后,得出顶骨厚度的标准偏差和厚度变化率较大这一特征规律.本算法对颅骨边界的描述准确,该方法将为建立人体头颅数据库、头部的生物力学造型等领域的研究提供参考依据.