Research And Applications Of Medical Volume Data Processing Algorithms In Digital Dentistry

With the popularity and rapid development of three-dimensional medical imaging technologies,medical volume data has been widely used in digital dentistry.Cone Beam Computed Tomography(CBCT),which is the most frequently used volume data source in digital dentistry,provides accurate three-dimensional information inside patients'heads.How to visualize CBCT data,and how to extract patients' personalized data models from CBCT data more effectively and efficiently,are two research hotspots in digital dentistry.The calculation and analysis of the relative relationship among patients'teeth,skull,and facial models provide significant references for the precise planning of dental treatment protocols.Focusing on the applications of CBCT data in digital dentistry,this dissertation is organized in three main parts:the visualization of volume together with mesh data,the extraction of the skull and facial models,and the extraction of the individual tooth model.Firstly,a hybrid rendering algorithm is proposed to visualize the volume and mesh data together based on the stencil test procedure in rendering pipeline.By combining the depth calculation of the two types of data,the proposed method makes the rendering result more accurate and the perception of the layered structure clearer.Secondly,based on a mixed threshold model and a self-adapted clustering model,an automatic method is proposed to extract the skull and facial models from CBCT data.The proposed method utilizes the global features of the image to efficiently achieve accurate segmentation results,and can automatically correct possible errors.Thirdly,a segmentation framework of volume data is presented based on the iterations of two-dimensional image segmentation.A hybrid level set based method and a regional energy function based method are proposed for single-layer CBCT image segmentation based on the structural characteristics of teeth,and the proposed framework improve the efficiency and accuracy of tooth extraction from CBCT data with these two segmentation methods.Finally,based on the volume data related algorithms above and the modular design concept,a digital dentistry prototype system for medical volume data is developed.This system provides support for personalized dental diagnosis and treatment by implementing the visualization of CBCT volume data and the extraction of teeth,skull,and facial models.