This study (This research is funded by Teranga Software and the ANRT CIFRE Grant no2019/1519 for CC.) proposes the use of Bayesian Networks for the prediction of unfavorable health events, and more especially pressure ulcers, in nursing homes. From a database of electronic medical records, we learn an explainable and relevant classifier, which performs better than the scores currently used in nursing homes.
Keywords: Bayesian Networks; Explainability; Machine Learning; Nursing Homes.