Distributed and real time data mining in the intensive care unit

Abstract

Pervasive computing, ubiquitous computing and ambient intelligence are progressively influencing health care and medicine. In this paper we present our work regarding the development of the INTCare System for decision support in Intensive Care Units (ICU). INTCare gathers data from several sources (e.g. bedside monitors, drugs system, lab results, the Electronic Nursing Record, etc) and is able to use that data to make predictions about organ failure and final outcome for each of the patients in the ICU. Moreover, the system continuously monitors its performance levels and is able to automatically adapt so that performance doesn’t drop over time.