Emergency departements (EDs) deal with high patient volumes, limited resources and time critical decisions. Triage is the first step in the ED and therefore, effective triage decisions are essential to ensure that patients receive the right care at the right time. However, mistriage remains a significant hurdle in the ED. The remEDy project first aims to develop a triage evaluation instrument that aids in identifying mistriage. We further aim to develop and evaluate machine learning approaches to identify patients at higher risk of mistriage and to support triage decision-making in the ED. This project will first review existing literature to provide an overview of current approaches and to identify methodological limitations and opportunities for improvement. Following this, existing clinical data as well as data from the remEDy project will be used to develop prediction models for identifying the likelihood of a patient being mistriaged while simultaneously highlighting the reasons for this assessment. Initial models will investigate well-established classifiers. More advanced approaches, which include multimodal models that can integrate different types of clinical information, would subsequently be explored.