Emergency departments (EDs) are highly challenging environments, characterized by overcrowding, limited resources, and a wide range of clinical severity among patients. In this context, triage becomes a fundamental step to establish priorities based on medical urgency, ensuring that the most severe cases receive timely care (TAHERNEJAD et al., 2024). Traditionally, triage systems use tags or color-coded cards to classify patients according to the severity of their clinical condition. Although these methods are well-established, they have significant limitations, such as limited space to record vital signs, difficulty in locating patients in high-demand settings, and a lack of accuracy in representing the patient’s real clinical status (TAHERNEJAD et al., 2024). Moreover, there is considerable reliance on the experience of the healthcare professional conducting the triage, which can introduce subjectivity and inconsistencies (DA’COSTA et al., 2025). In this scenario, artificial intelligence (AI) emerges as a strategic tool to transform triage processes. The integration of techniques such as machine learning (ML), deep neural networks (DNNs), and natural language processing (NLP) contributes to the standardization, accuracy, and efficiency of triage in EDs (DA’COSTA et al., 2025). AI-based solutions can operate in an automated and integrated manner, from collecting and analyzing clinical data to providing real-time feedback, supporting clinical decision-making through robust algorithms. Recent studies also highlight the emerging role of AI-assisted virtual triage. This approach involves the use of remote electronic tools in which patients themselves enter their symptoms into automated platforms, which then assess the severity of the cases and guide the type of care needed (NASSER et al., 2025). Virtual triage has proven to be viable and promising, especially in high-demand scenarios such as during the COVID-19 pandemic, reducing individual biases and improving the accuracy of clinical recommendations