Machine learning methods for hospital readmission prediction: systematic analysis of literature

Hospital readmission is one of the challenges that force an extra pressure and financial burden on healthcare and causes a significant waste of medical resources. However, some of these readmissions could be predicted and preventable. For this prediction, identifying the patients with high readmission rates is necessary before discharge to make appropriate interference to impede the readmission. Using smart technologies, and their collected data help in preparing a large amount of medical data sets suitable for Artificial Intelligence and machine learning to extract data insights and trends. Recently, there has been a significant interest in predicting readmission using artificial intelligence including machine learning methods. However, most of these studies focus on specific aspects of the prediction process and very few provide a comprehensive machine learning process in readmission prediction. Therefore, the objective of this article is to provide a comprehensive review of the recent studies on machine learning algorithms. In addition to the systematic literature review, by integrating the contribution of previous studies we also present the findings in a framework to cover all stages of machine learning for predicting the chance of hospital readmission. © 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG.