Research On 3D Reconstruction Of Medical Image Based On Improved MC Algorithm

With the rapid development of interdisciplinary fields such as computer science and graphics,the promotion of computer graphics is changing with each passing day.In the development of such disciplines,the development of medical images is particularly prominent.Three-dimensional reconstruction technology is one of the core hot topics in the field of medical image research.This technology aims to extract two-dimensional tomographic image from three-dimensional scalar data field and convert it into three-dimensional image,which compared with the two-dimensional tomogram,it is more stereoscopic and real.In the field of medical images,it can be analyzed more intuitively and directly to the lesions to avoid misjudgment.After comparing the three-dimensional reconstruction surface rendering and volume rendering methods,this paper selects the research method of surface rendering because of the high precision and real-time characteristics of the surface rendering method.The MC(Marching Cubes)algorithm is the most classic in the surface rendering methods.Therefore,after the detailed analysis based on the research status of the MC algorithm,in order to improve the MC algorithm,due to the ambiguity of the cube connection,it is easy to generate holes and it's drawing speed is slow.A method based on improved MC algorithm for 3D reconstruction of medical images is proposed.First,the image is segmented to extract useful areas before 3D reconstruction.This part is done in Mimics.The processed image is reconstructed in three dimensions,and the improved MC algorithm is used for 3D reconstruction.In order to prevent voids when the cube is connected,the original 15 topological configurations in the MC algorithm are increased to 24,which effectively avoids the generation of holes.In order to avoid the problem of slow calculation speed,this paper adopts the midpoint selection method instead of the linear interpolationmethod,and divides the 24 configurations into three types,each type corresponding to one thread,and further improves the calculation speed through multi-thread parallel processing.Multi-thread synchronization is implemented by using the critical section,and then a protocol mapping table is designed according to the idea of the message mapping table.The function pointer is triggered by the macro,and the processing function is called by the function pointer to complete the encapsulation of the protocol mapping table,and the class' s Open Closed Principle is maintained,and guarantees the scalability of the class.Afterwards,the accuracy and speed of the algorithm are verified and analyzed by comparison.The speed verification verifies the accuracy of Windows and Linux respectively,it proves that the algorithm speed is improved by nearly 40% under the premise of ensuring accuracy.Finally,the algorithm is applied to the software of medical image 3D reconstruction.The software is implemented in C++ based on the MFC framework.The applicability and accuracy of the algorithm on medical images are demonstrated by two sets of examples.