Association models for relating problems with semiologic data in intensive medicine

Abstract

In Intensive Medicine, the large amount of data that medical professionals are subject to can be overwhelming, leading to the use of techniques and treatments that may not be the most effective in treating patients. Should there be a need to cross planning registries made by doctors and nurses with patients' problems, the situation becomes unmagenable. To support health professionals' decision-making process, and consequently allow health professionals to make informed and timely decisions, by promoting proactive actions, the current study approaches the establishment of a correlation between medical problems and medication and therapies, using association rule mining algorithms, so that physicians can have the correct and timely information regarding patients and consequently, the most appropriate treatments for them in every situation. The main objective is for doctors and nurses to be able to look through problems and have them associated with the most frequently used and reliable therapies and medication, in order to assist patients with the highest healthcare quality. The results of this work corroborate that in order to improve the care provided to Intensive Care Units patients, it is essential to implement intelligent systems that can support hospital staff and assist to provide healthcare more efficiently.