Research On Image Processing And 3D Reconstruction Of Orthopedic CT Data

According to the statistics of medical data of Ministry of health,orthopedics diseases have seriously endangered the health of citizens of our country.The incidence of various kinds of orthopedics disease is increasing year by year.The traditional way of diagnosis and treatment is to diagnose the patient's condition by directly observing the CT image.This way is easily influenced by the image quality and the experience of doctors.The digital image processing technology is used to denoise,segment and reconstruct the CT images in orthopedics to generate three-dimensional geometric model,which can further improve the accuracy of diagnosis.This is conducive to the clinical diagnosis and treatment of patients of orthopedics disease.In this paper,different types of orthopedics diseases image sequences are preprocessed and 3D reconstructed.In filtering,a hybrid filtering algorithm based on adaptive template is proposed in this paper.The image is divided by using different templates,and the gray standard difference and rough granularity of the corresponding local regions are obtained,thus the best threshold is determined and adaptive filtering is made for the local regions with different features.In segmentation,this paper proposes an improved TV-L1 segmentation algorithm.The dual variable and main variable are introduced in this method.At the same time,the fast principal dual method is used to convert the traditional TV model to the semi implicit iterative calculation of the dual variable and the exact calculation of the main variable.The fixed iterative method is applied to obtain the minimum solution of the energy functional.This method can solve the problems of texture information loss and excessive segmentation in the traditional TV model processing orthopedics images,thus further improving the accuracy of the image segmentation of the pixel distribution inhomogeneous.In the reconstruction process,in order to help the orthopedics experts to observe the patient's pathological structure from various angles and all aspects,the three-dimensional reconstruction of the two-dimensional image sequence is made by using the moving cube method and the ray projection method respectively,thus creating a three-dimensional geometric model.Finally,we design and develop a computer-aided diagnosis and treatment system which can achieve import and export,preprocessing,surface rendering,volume rendering and interactive operation of image data.