An Efficient Strategy for PACS

Conventional PACS systems are designed based on centralized data repository architecture, for integrity and security reasons. This design is highly sensitive to escalation in number of retrieval and storage operations, which represents a major limitation that reduces the efficiency of PACS used in applications that require repetitive data access e.g. medical education and medical decision support systems. In this work, we propose two techniques to overcome this limitation, without changing the PACS architecture. The first one, is based on reducing the number of data requests on PACS archiving server using a virtual distributed system; the second one, is based on redistributing data requests over time, using a prediction system. Simulation results show significant improvement in performance up to several orders of magnitude compared to conventional PACS design.