부산대학교 도서관

there is currently a significant amount of technology in hospitals in particular in the Intensive Care Units (ICU). The clinical data daily generated are integrated into Decision Support Systems (DSS), in real time for a better quality of patient care.The hospital environment has many outbreaks of infections, objects or environments in which microorganisms can survive or multiply, such as the facilities, invasive devices or equipment used, or even patients, health professionals and visitors. The existence of nosocomial infection prediction systems in healthcare environments can contribute to improving the quality of the healthcare institution. It also can reduce the costs of the treatment of the patients that acquire these infections. The analysis of the available information allows preventing these infections which can help to identify their future occurrence. This paper presents the results of applying models to real clinical data. Good models were obtained, induced by the Data Mining (DM), K-Means and K-Medoids Clustering techniques (Davies-Bouldin Index 0.14). These models, classification models, should act in a DSS capable of helping to reduce this type of infections as well as reduce the costs associated with them.