The Key Technology Research Of3-D Reconstruction In Medical Image Based On MC Algorithm

The purpose of Medical image3-D reconstruction is construct3-D medical image geometric model by treating2-D image sequence of CT and MRI. But now there are som problems such as memory space is large, reconstruction speed slow, reconstruction existing debris and other issues in3-D reconstruction. To solve these problems, this paper made the following research:The data process for sectional medical image (CT images). It gets a DCM file interpretation method used to withdraw the required information for the3-D reconstruction of medical imaging by analysising the standard of the file DICOM DCM format and DCM file data sets structure. And then, combined with the idea of run-length coding and block coding, it can make the compressions store of image data quickly in memory. The coding method for DICOM image, not only save memory space and improve transmission efficiency, but also read the required part of a single pixel, or pixel data from the compressed data conveniently.The pretreat befor the3-D reconstruction of medical imaging. First, hybrid median filtering is adopted to remove noise, and it can keep fine image features intact. Second, structure gets by using the one-way linear interpolation method and building3-D data field through the DICOM sectional sequence structure is more facilitate for3-D reconstruction. Again, seeded region growing algorithm is used for segmenting object tissue, and the segmentation result can convenient to compressed and written into segmentation matrix.The3-D reconstruct of medical imaging. The process of the3-D reconstruction of medical imaging is process of extracting algorithm. This paper discusses the present medical image used in the standard of the principle and the Marching Cubes (MC) algorithm of the shortcomings of the existing is discussed in detail, improved the traditional MC algorithm in two major ways:One is to adopt the triangle surface modification connected sequence look-up table method to eliminate the topology of ambiguity in standard MC algorithm of existing; another is storage the image data by3-D array way, and establish look-up table to record each configuration mode of the corresponding adjacent body, and introduce a queue to record the body element with the method to improve the calculation efficiency of MC algorithm.The improvement of the MC algorithm reconstruction of3d model, adopt debris removal algorithm, to eliminate the reconstruction process because of floating point calculation error and produce pieces. After defragmentation, the data of3-D mesh model can greatly reduce, and the efficiency for the follow-up rapid prototyping can improved greatly.