In the present, the applications of ultrasound imaging for medical diagnosis become more and more important. But the noise and spots are often existed in the ultrasound images, and affect the quality of images. In clinical, these will affect the final diagnosis of the physicians. Now, 3D ultrasound become more and more popular, these problems have to find an efficient method to solve. This study will focus on the 3D reconstruction problem from 2D continuous ultrasound images. We also have to discuss the problem of noise and spots, and find some effective methods to overcome.In the proposed system, we proposed some procedures to handle the 3D image registration and reconstruction. In the 3D reconstruction, we focused on the tumor segmentation and reconstruction by the triangle grids. The experiment results show that the performance of this system is satisfactory.
關聯:
Journal of Image Processing and Communication 1(1), p.41-48
