Research On Medical Image Segmentation And 3D Reconstruction

The main work of this paper is brain tumor image segmentation based on Chan-Vese model, the 3D reconstruction and position of medical image and the calculation of the liver tumor volume.First, this paper describes the background and significance of the brain tumor image segmentation,3D reconstruction and the calculation of the liver tumor volume. On that basis, the commonly used image segmentation methods are introduced and classified.Secondly, this paper introduces the Chan-Vese segmentation model and proposes the brain tumor image segmentation method based on Chan-Vese model. The brain tumor image is first eroded iteratively by using morphological erosion to extract the brain tumor contour. Then the brain tumor included in eroded image is segmented by the Chan-Vese model. The morphological reconstruction is then used to optimize the segmentation result. The image is finally dilated iteratively by using morphological dilation to restore the image. The proposed method could reach a better segmentation result than the Chan-Vese model When the boundary gray value of the segmentation target change slowly or the segmentation target includes big corner and narrow-band region, in this way this method could enhance the accuracy of the image segmentation of some brain tumor.Thirdly, all segmented brain tumor images are reconstructed and located to three dimensions. The goal of 3D reconstruction technology is to reconstruct 3D information of the object from the 2D image slices, so medical professionals could observe and analyze the lesion body from different angles. In this paper, on the basis of introduction of the surface rendering and volume rendering, we use the visualization toolkit VTK to reconstruct and position the brain tumor from the segmented 2D medical images, the result could provide a theoretical basis for the doctor's preoperative analysis.At last, we make a scientific calculation on liver tumor volume. As the liver tumor volume could help medical professionals to understand objectively relevant characteristics of liver tumors, so the calculation of the liver tumor volume has a great application value. In this paper, we use adding method and VTK-based calculation method respectively to calculate the liver tumor volume, making a statistics on experiment results.