The Analysis And Preprocessing Of DICOM Format Cerebral Images

With the rapid development of computer science and medical imaging technologies, many kinds of new medical imaging methods are emerging for extensive applications in practice. At present, most medical information equipments, including the imaging equipment and the information system in hospital, follow DICOM standard gradually. To analyze DICOM, collect the information and the data in DICOM. read the medical digital images and then preprocess them can further help the doctor diagnose the pathology and the illness. As a result, the basic step of developing the medical software is to analyze and utilize DICOM and to read the medical digital images and then preprocess them. It illustrates a good prospect of application.The thesis mainly analyzes DICOM. use integrated software development tools to parse and preprocess the CT images. The following tasks have been carried out.Firstly, by carefully reading Digital imaging and Communications in Medicine 3.0 standards, the organization structure and the encoded mode of DICOM files is analyzed and understood. The specific resolving process and method is proposedSecondly, through the parsed CT data and the equipment parameters, the basic operation such as zooming, rotating, pseudo-coloring and sharpening are processed. The result of the operation can be saved as the general image format of BMP. In order to realize the three-dimension reconstruction, the CT images should be interpolated. The thesis analyzes the interlayer interpolation, and by taking the practical application and the efficiency into consideration, the linear interpolation is adopted, which the image data can be viewed and displayed from cross section, sagittal section and crown section. Then the essential auxiliary diagnose information can be provided for doctors by computing the distance, angle and area of CT images.Thirdly, the display technique of CT image data is studied. Because the pixel dimension of the medical image taken from the instrument is high, the original image data should be mapped into grey-based image for display. And the different medical structure can be further observed through the technique of window width and window level. Color are more sensitive than grey for people's eye observation. The ability of people's eyes to distinguish the medical image can be improved by means of pseudo-color transformation for medical images. The thesis chooses HSI color space to do pseudo-color in order to show the greater grey scales. Combining multi-window width and window level and pseudo-color the better display results are obtainedFourthly, the DICOM CT image preprocessing experimental system has been designed and implemented. On the basis of the former analysis, the function of the system are designed, and C# language and VS integrated development tools are used to realize the preprocessing module function. The experimental data are provided by Anhui Provincial Hospital. The system can correctly analyze and display the CT image data and realize the designed image preprocessing functions.