Wavelet Analysis And Its Application Of Mammography Analysis

The early detection of breast cancer is clearly a key ingredient of any strategy designed to reduce breast cancer mortality. Therefore, it is very important to analyze mammography by means of computer since the early detection of breast cancer mainly depends on mammography. Improvements of the visibility of some features of interest ( for example, lump, calcified tissue ) with the help of the enhancement techniques play an important part in recognizing interesting features in mammography as early as possible. Thus, the enhancement techniques in mammography analysis have been extensively studied.The enhancement of image has linear and non-linear enhancement, and includes global enhancement and local enhancement. Global enhancement changes all pixel in image; local enhancement only changes pixel of some important feature in image. Multiscale edges identified within distinct levels of transform space provide local support for image enhancement by multiresolution. Mammograms are reconstructed from wavelet coefficients modified at one or more levels by local and global non-linear operators. In each case, edges and gain parameters are identified adaptively by a measure of energy within each level of scale-space. We show quantitatively that transform coefficients modified by adaptive non-linear operators can make more obvious barely seen features ofmammography without requiring additional radiation. In addition, We demonstrate that features extracted from multiresolution representations can provide an adaptive mechanism for accomplishing local contrast enhancement. By improving the visualization of breast pathology we can improve chances of early detection.The paper shows that wavelet transformation coefficients, modified by non-linear operators, can reconstruct mammography and make more obvious barely seen feature (mass) of mammography in order to detect breast cancer more early and accurately. Among wavelet transformation we make main use of biorthogonal wavelet transformation with filter length 5 for global enhancement and B-splines with compact support for local enhancement. We implement the methods by programming two software packages. One was programmed in VC++ and MATLAB, the other was programmed in VC++. The two software packages are marked by good user interface and obviously processed effect. The result of the experiment shows that global enhancement makes some key feature more obvious in mammography, and local enhancement makes the details of some important feature more obvious in mammography. Our result are compared with other wavelet image enhancement techniques by measuring the local contrast of known mammographic features.The structure of this paper is that the first chapter introduces harms of the breast cancer, mammographic analysis, the summary of wavelet ,my work and experimental result; the second chapter discusses the multiresolution analysis and wavelet analysis; the third chapter discusses methods of the mammographic enhancement; the fourth chapter introduces software packages and experimental result in my experiment; the fifth chapter summarizes my experiment and presents the direction of the study of mammographic enhancement.