In health systems, there has been an emergence of new types of data and new technologies that allow continuously monitoring the status of the patients and make easy the achievement of real time information. The storage of all the acquired information makes possible to identify trends in medical data by means of new Clinical Decision Support subsystems. Current knowledge management solutions are specific, proprietary and closed and can not perform real-time analysis to improve the patient's diagnosis or treatment. There are neither solutions that integrate the large amount of heterogeneous information that nowadays are available in health environments. To overcome these objections, this paper proposes a new approach to design a data management system in a telehealthcare system with specific real-time constraints on knowledge acquisition and generation. It is a preliminary study and presents the main features of the system architecture and a preview of the technological solution implemented.