DISTRIBUTED REAL-TIME LOSSLESS IMAGE COMPRESSION OF MEDICAL IMAGES

Abstract

Remote diagnosis of patients by a specialist over vast distances is becoming a global reality. This necessitates the transfer of very large files containing the patient's data which may include video, still images and sound. It is a challenge to get the information over long distances efficiently and cheaply. This calls for image compression (specifically for images) to reduce the file size of the data before transmission from the remote site. Medical images demand lossless compression for diagnostic and legal reasons which is less efficient than lossy compression in terms of compression ratio. The compression of videos requires fulfilment of real-time requirements at 25 fps. This can be achieved by hardware or software. Software is preferable due to its lower cost and its efficiency can be boosted by working in a distributed computing environment. This report covers the various steps involved in the implementation of a distributed real-time lossless image compression, specifically for MRI images. The implementation required the selection of the components (hardware and software) for a distributed computing environment bearing in mind the setting of its potential use. The FELICS algorithm was modified to better suit a distributed environment. The distributed system is approximately 15% faster than a localised system but still falls short of real-time requirements.