Medical Recommender Systems: a Systematic Literature Review

Medical recommender systems are applications in the field of health. These systems use Artificial Intelligence techniques to provide personalized recommendations to healthcare professionals and patients, based on available and relevant patient information. Software engineering is essential in developing medical recommender systems, as these systems must be accurate, reliable, and secure for use in clinical settings. This work presents a Systematic Literature Review based on the Kitchenham and Charters guide, in order to explore the Artificial Intelligence techniques used in this type of system, which can be incorporated or improved by software developers who participate in this type of project. Twelve primary studies were selected, where mainly machine learning approaches were identified (algorithms based on decision trees, neural networks, Bayesian classifiers and clustering such as k- means), matrix approaches, based on rules, among others. Precision, Recall, and Root Mean Square Error (RMSE) were the main measures used to evaluate the performance of these systems. Finally, the studies propose always increasing the sample size of the tests carried out, including relevant patient information such as social networks and clinical information, as well as exploring other algorithms and approaches that allow improving the results of the recommendation.