Framework for Medical Data Compression in Telemedicine

Telemedicine as a set of systems and services that enable the sharing of medical diagnostic imaging data remotely is an important factor in the achievement of improving overall health system and services. Medical imaging generates large amounts of data. An MRI study can contain up to several gigabytes (GB). The obtained image data together with other metadata (about the patient) packed in standardized formats such as DICOM and NIfTI are stored in a centralized data repository. From there the needed data are sent to the client device (PC) and presented to a specialist who performs diagnostics. The exchange of such large amounts of data in the local network facilities is a significant problem due to bandwidth sharing which is even more significant in mobile and wireless networks. A possible solution to this problem is data compression with the requirement that there is no loss of data. This work presents a novel framework for four-dimensional medical data compression architecture. This framework is based on different procedures and algorithms that detect time and spatial redundancy in recorded MRI volumes. Motion in time is analysed through motion estimation based on neural networks. Motion estimation is used to eliminate a large amount of temporal and frequency redundancies that exists in sequences of 3D data. Combination of segmentation, block matching and motion field prediction along with expert knowledge are incorporated to achieve better performance. Spatial analysis is done through an extension of wavelet transformations to three dimensions. For still volume objects different wavelet packets with different filter banks provide a wide range of frequency analysis. The suggested data compression architecture incorporates operations for spatial and time analysis of datasets, creating kernel shapes and models, and fitting of medical 4D datasets. With combination of removing temporal and spatial redundancies, very high compression ratio can be achieved.