A Semantic-Based Platform for Medical Image Storage and Sharing Using the Grid

Abstract

Since the introduction of medical imaging techniques such as MRI, X-Ray, SPECT, PET, large amounts of medical images have been produced in the different fields of medicine. Most of this data is usually stored either in paper format (stored in clinical records) or in optical disks (given to the patients) and this information is partially uncorrelated with the clinical history of patients. Furthermore, the image processing carried out by the physicians that perform the exam and used to make the diagnosis is unknown for other researchers that further treat the same image. In this paper we propose a system that allows radiologists to share medical images (and their processing) and the associated metadata both within the same medical institute and with other medical institutes located anywhere in the world by using GRID services for data (LFC) and metadata (AMGA) storage. The system is also provided with a semantic layer for describing the stored images through a novel RDF schema, which integrates existing ontologies and vocabularies such as FOAF, Mesh and GeneOntology with new terms related to the image processing part. This enables the correlation of medical images with other clinical information and makes our system fully compatible with the existing systems compliant to semantic web standards.