Study On Key Technologies Of The High-resolution Medical Ultrasoud Imaging Workstation

In recent years, with the coming of information age especially digital age, Picture Archiving and Communication System(PACS) has been devel-oped rapidly, and digital imaging diagnostic modes have become an important aid for medical diagnosis and treatment. Ultrasound medical imaging technol-ogy plays an important role in medical diagnosis, surgical planning, treatment, postoperative monitoring and other links for it's safe, fast, real-time and low-cost.Early medical imaging equipment produced images in different formats. Hospital's different departments have different image standards. Even different regions differ. Diagnostic images can not be passed between different segments. This directly affected the patient's diagnosis and treatment, and also resulted in slowing the development of medical research. To this end, American College of Radiology and the American National Electrical Manufacturers Association launched Digital Image and Communication Standards in Medicine in 1996. This standard is an agreement based on TCP/IP about the general medical image file format, communication and print. Its 3.0 standard is the first glob-ally accepted standard of medical digital imaging communications. In recent years, it has been widely promoted and applied in our country.The ultrasound image based on DICOM standard, due to its imaging mechanism, has a low resolution. Currently we mainly rely on clinical experi-ence in medical diagnosis, so the accuracy needs to be improved and the de-mand for high-resolution ultrasound imaging workstation is now urgent. Med-ical image processing technology as the foundation of the development of these imaging technologies, drives profound changes in modern medical diagnosis.In this paper, we use the characteristics of medical ultrasound images combing with image processing technology to research deeply on the high-resolution medical imaging techniques, including image denoising, edge en-hancement and image magnification etc.Firstly, we analyze the background of the medical ultrasound imaging workstation, and explain the importance of the ultrasound imaging workstation and the practical significance of this issue.Secondly, we give a introduction on related knowledge of the medical ultrasound imaging workstation, including principles of ultrasonic imaging, ul-trasound image format, the characteristic of ultrasound images, etc.Then, we study the technology of the medical ultrasound image processing on the basis of the analysis of a large number of medical ultrasound images and the method of traditional image processing. After that we introduce some suit-able methods for ultrasound image processing, including noise removal, edge sharpening and amplification methods, etc. On this basis, we propose two new algorithms with high-resolution ultrasound imaging workstation requirementsUltrasound image denoising and edge enhancement. We found that the choice of the diffusivity function is generally a function of|(?)u| in a lot of liter-ature for the diffusivity function. However, during image processing, we hope to combine anisotropic with isotropic diffusion equation's models, and make a image selective proliferation in different regions. Therefore, we make g(s) function optimized, propose a derivative-driven anisotropic diffusion equation, and make a research for ultrasonic image denoising and image enhancement.Ultrasound image enlargement and edge enhancement. Classical amplifi-cation method, is just make a image pixel interpolation, in many cases, will obscured the image details. It could not remove noise and does not meet the requirements of high-resolution ultrasound imaging workstation. Therefore, we combine the two-way diffusion equations, and use the second order directional derivative to limit the diffusion rate. We propose a amplification method on two-way diffusion of ultrasound images.We do a lot of experiments by using clinical ultrasound images and ad-justing parameters of two types of algorithms, we have obtained better image processing results.Finally, we discuss the implementation of software. Visualization process is a very important aspect of ultrasound workstation. In the Matlab plat-form, we can designed a visual program to run on different operating systems separately by using already-designed image processing algorithms. That can provide an image processing interface, human-computer interaction, to im-prove the function of high-resolution ultrasound imaging workstation. In the software, we combine clinical practice requirements. Demanding for different sections of the software development, it will be more practical value. And this is the future direction of software development.This thesis comes from the following research projects:The research of shock diffusion equations and numerical methods of diffusion equations in image processing(Fund of key Laboratory of National Laboratory of Pat-tern recongnition(NLPR), Chinese Academy of Sciences State, NO.2006-8); Nonlinear shock diffusion equations and numerical methods in image process-ing(Natural Science Foundation of Shandong Province, NO. Y2006G08):Med-ical image processing method of partial differential equations and numerical computation (Teaching and Research Fund of Institute of Mathematics, Shan-dong University, NO.200809); Image restoration based on image decomposi-tion and adaptive regularization method for numerical calculation(First pro-viding financial aid of China Postdoctoral Science Found, NO.20090460089).