Peer-to-peer data discovery in health centers

Abstract

The sharing and integration of health care data such as medical history, pathology, therapy, radiology images, etc., is a key requirement for improving the patient diagnosis and in general the patient care. Today, many EPR (Electronic Patient Record) systems are present both in the same or different health centers and record a huge amount of data regarding a patient. In most cases the care treatment of a patient involves different healthcare facilities, including the cares provided by the family doctors. Managing these data, typically petabytes or terabytes in size, and optimizing the applications (image analysis, data mining, etc.) for these architectures is one of the challenges that must be tackled. Therefore, there is a clear need for the design and implementation of new scalable approaches to deal with the associated information overload and cognitive complexity issues. A possible solution involves considering a simplification of data coming from different EPRs, in a structured schema, typically called a meta-EPR. Owing to the security of patient data, each health center manages its own meta-EPR whereas a framework integrates these data among different sites. This work addresses the issue of sharing and integrating health care data, proposing a meta-EPR, based on Peer-to-peer (P2P) technology for data fusion. We describe an implementation of a distributed information service, that shares meta-EPRs and provides aggregation of relevant clinical information about patients based on a structured P2P overlay.