Development of Algorithms for Medical Image Compression

Abstract

Image compression algorithms are developed mainly for reduction of storage space, easier transmission, and reception. In this chapter, many image compression algorithms have been developed based on various combinations of transforms and encoding techniques. This research work mainly deals with the selection of optimum compression algorithms, suitable for medical images, based on the performance indices like PSNR and compression ratio. In order to find the effectiveness of the developed algorithms, characterization of the CT lung images are performed, before and after compression. The diagnosis of lung cancer is an important application for various medical imaging techniques. In this work, optimal texture features are identified for classification of lung cancer have also been incorporated as a case study. The texture features are extracted from the in CT lung images. BPN is trained to classify the features into normal and cancer.