Research Of The Thickness Measurement Of Skull Based On Double-Snake Model

Living human skull measurement is applied in many researching domains, such as, anthropology, designing of medical machines, biomechanical modeling of head injury, et al. CT(Computed Tomography),a medical imaging technology, makes it is possible to measure living human skull by computer autoly. At the same time, digital image segmentation and boundary extraction is a very important subject of computer vision, and it is a non-lack sector of many researching domains. So applying a image segmentation algorithm on a CT image of head is a very useful method to measure living human skull. Active Contour Model is a image segmentation algorithm which is paid great attention in recent years, it works very well on medical images. The task of this thesis researches method of living human skull measurement based on CT image by using snake model.This paper makes some improvement on the traditional snake model to a double-snake model which can be fix its initialization automatically and extracts the inner and outer boundaries of the goal object at the same time. This model can extract the inner and outer boundaries exactly and quickly at the same time. We use this model to extract the skull boundaries on a particular CT image. Then propose a method to measure each bone of skull thickness and other characteristics by using the trait of the boundaries former. Then we get the characteristics that standard deviation and the rate of change of the thickness of parietal and occipital bones are larger.To measure skull thickness based on one CT image has some limitation cause of this image should be a particular one.3D-rebuilding is a very important technology which means rebuilding a3D modle from a series of2D images. We extract the boundaries of a series of CT images exactly by using double-snake model. Then based on the trait of the boundary, we use a sample, efficiency method of3D rebuilding to build a skull model. Then we proposed a method to measure skull thickness by finding a point and a face with a relative relationship. This method and the method of base on single CT image both can provide a reliable basis for those applications.