Application of data mining for the prediction of prophylactic measures in patients at risk of deep vein thrombosis

Abstract

In the last decades, with the increase in the amount of data stored in the healthcare industry, it is also extended the possibility of obtaining important information to support the decision-making process of health professionals. This article has as evidence to apply Data Mining (DM) techniques to health databases of patients with medical Deep Vein Thrombosis (DVT) risk, with the objective of classifying, based on different attributes obtained in medical discharge reports, the main prophylactic measures taken. Therefore, to achieve this goal, the free software Weka was used aiming to facilitate the process of DM, along with the algorithms chosen. In view of this, it was concluded that the service to which each patient is associated is the most relevant factor for prophylactic measures followed by the age range to which the patient belongs. This study also deduces that it can be possible to obtain classifiers capable of predicting the best prophylactic measures with a qualitative level similar as one of a health professional and, thereafter, it can be possible to obtain the classification.