Ensemble classifiers in a serious game for medical students in clinical cases.

In recent years, medical students from Brazil have obtained poor performance in the CREMESP (Regional Medical Council of Sao Paulo State) exam. Although the most recent assessment shows improvements, they still perform poorly in specialties such as clinical cases, diabetes, and myocardial infarction. In this context, several strategies emerge to improve the teaching and learning of medical students, one of which is serious games. However, there is a problem with trying to train medical students in clinical cases through a game because there are so many diseases, and processing a database containing those diseases is laborious. This paper describes an intelligent module of a serious game for training medical students. This module has a website for data dissemination, management and expansion. The information is made available for serious game through a web service. We used databases of a clinical case, heart disease and diabetes. These databases are rated by the machine learning ensemble classifiers. The smart module was validated by F-measure using percentage division and cross-validation. With the results found, it was provision that the intelligent module can manage the of serious game data, because it has a whole structure for this purpose.