Interactive pattern recognition in cardiovascular disease management. A process mining approach

Abstract

Cardiovascular diseases are one of the main causes of mortality in the European Union. Cardiovascular diseases require a continuous follow-up in order to prevent complications that produce a decrease on the quality of life of patients and can even lead to their death. According to the Evidence Based Medicine thesis, the involvement of the patient in the care process will improve the Quality of Care provided and the efficiency of Care Plans. However, there is controversy in the literature about the fact that the Care Plans that Evidence Based Medicine provides are based on strict clinical cases to generate scientific evidence that is unclear whether they can be applied to personalized medicine. The use of an Interactive Pattern Recognition approach can create models controlled by human experts based on the biomedical data provided, in addition to medical decisions and the patient personality, providing a better way to find more tailored models. However, the use of Interactive Pattern Recognition techniques still requires the application of Human Understandable Pattern Recognition techniques. In this paper, an Interactive Pattern Recognition technique based on Process Mining technology is presented as a solution to deal with this problem.