Redundant Wavelet Transform in Biomedical Image Application
暂无分享,去创建一个
The change of blood vessel is very important in the biomedical research, because creatures, including human beings tempt to adjust themselves to fit for the new environments. For example, if a tumor was found inside a human body and diagnosed as cancer, to get rid of this tumor is important, but to prevent the tumor to reach out to other organs is more crucial. The goal of starving the tumor can be achieved by preventing new blood vessels to grow into it. Another example is the wound healing which can be enhanced by encouraging blood vessels to reach the damaged parts. Both applications focus on the trace of the growing blood vessels. There are several trials on how to trace blood vessels, we can category them into three main approaches: window-based methods, classifier-based methods and tracking-based methods [1]. In this paper, the authors explore the usage of RDWT in this specific area of biomedical image application. Redundant wavelet transform (RDWT) has the property that all of the RDWT coefficients locate at the same spatial location as in the original spatial image, which makes it preferable for feature extraction and signal analysis [2]. With the already successful applications of RDWT in image processing and video coding [3,4], the authors built a new system to segment blood vessels using RDWT coefficient mask.Copyright © 2006 by ASME