RECARGA: Soft computing for gamification analytics in cardiac rehabilitation

Cardiovascular diseases are the leading cause of death in Western countries, causing 4 million deaths per year in Europe. The aim of RECARGA is to use gamification techniques to facilitate the execution of Cardiac Rehabilitation (CR) Programs, which are intended for patients who have suffered a serious cardiovascular episode. CR programs include multifactorial processes and aim to ensure optimal physical, mental and social conditions for those patients who have suffered a coronary event, allowing them to take a normal position in the society by their own means.

Starting from the premise that gamification techniques are aimed to increase motivation, engagement, effort, and other positive game values, it seems appropriate to apply these techniques to promote a healthier lifestyle among patients and encourage their wellness. To achieve this objective, we will personalize the CR programs to the characteristics of each patient through the gamification.


The main objective of the project is to develop a set of algorithms, based on soft computing techniques, for the automatic discovery of information about the behavior of patients who follow RC programs. By means of these algorithms we expect to improve the design and personalization of RC programs, adapting them to the specific characteristics of patients at any time.

The specific objectives of the project are the following:

  1. To develop process mining algorithms for the automatic discovery of the real flow of activities followed by the patients of CR Programs.
  2. To develop algorithms for the prediction of the time required in CR programs, with the aim of helping doctors to establish the temporal restriction of the execution of activities and plans.
  3. To develop algorithms for the prediction of exceptions and of patients progress, with the aim of discovering possible conflicts and anomalous situations in the behavior of patients following CR programs.
  4. To Integrate the information extracted by the machine learning algorithms in the CR decision support system and thus help physicians to better personalize the treatment/program of patients