Fleuren LM, et al. - Researchers undertook a systematic review and meta-analysis in order to assess the performance of real-time models to predict sepsis that have emerged with the advancement of machine learning. Searching PubMed, Embase.com, and Scopus systematically, they identified 28 papers eligible for synthesis. From these papers, they extracted 130 models. Development of the majority of papers was done in the intensive care unit (n = 15; 54%), followed by hospital wards (n = 7; 25%), the emergency department (n = 4; 14%) and all of these settings (n = 2; 7%). Findings support that on retrospective data, individual machine learning models can assist in making an accurate prediction of sepsis onset ahead of time. Although these models present alternatives to traditional scoring systems, the assessment of pooled results is limited by the between-study heterogeneity.