In the health care industry, the management of ever-increasing medical data becomes a real challenge. The development of imaging technologies, the long term retention of medical data imposed by medical laws and the increase of image resolution, are all causing a tremendous grow in data size. As a consequence, the cost of a medical exam has significantly increased, storage cost being about 75% of the total cost (Yansys: yansys-medical.fr).
Medical data management thus leads to scalability, maintenance and performance issues. The research challenge we tackle in this paper is as follow: is it possible to build a system that: (1) is highly available, (2) is cost-effective, (3) enables to store huge/ever increasing data, (4) provides high expressiveness/ad-hoc queries and (5) supports heterogeneous medical images (different specialties, modalities and physicians practices).