Study Of Medical Image Compression Algorithms

With the development of modern medical technology and the widely use of PACS (Picture Archiving and Communication System) in medical devices, it is important to solve the archiving and transmission problem of medical pictures in PACS. The key is how to realize the effective compression of medical pictures.Integer wavelet transforms (IWT) based on lifting scheme are introduced systemly in this paper. In this part of the paper, the relationships of IWT and image compression are pointed out and several formulas of IWT are listed. This paper compares the nature images and marrow cell images statistically and includes some characters of marrow cell images. In the application of marrow cell images compression, (4,2) transform is found to be the best by analysis of energy distribution, percentage of zero coefficients and entropy of wavelet coefficients. The principle and programming of embedded zerotrees wavelet coding are introduced in detail. A new method of quantization threshold is applied to EZW coding. The experiment results show that this approach can increase the number of zerotrees and improves marrow cell images compression.At last, this paper introduces a platform of cell images database and discusses design technology. In order to demand for medical diagnoses, some useful functions such as search, images processing and analyzing are designed except tradition database functions in this platform system. Functions and data of the platform can be conveniently extended for different applications.