AI Symptom-Checker Could Help Emergency Doctors Prioritize Patients
Key Points:
1. A new AI-powered triage system was shown to outperform junior clinicians in identifying patients requiring emergency care across multiple clinical scenarios.
2. The model accurately distinguished high-urgency cases with improved sensitivity and shorter assessment times, suggesting potential utility in emergency department (ED) workflows.
Emergency departments (EDs) continue to face mounting challenges related to overcrowding, long wait times, and increasingly complex patient presentations. Researchers at KTH Royal Institute of Technology and the Karolinska Institute developed an AI-powered tool designed to support triage by estimating the urgency of cases based on symptoms and vital signs. In a multicenter study across three Swedish hospitals, the system was validated in nearly 3,000 real-world ED cases. The model achieved a sensitivity of 84.8 percent and specificity of 76.1 percent in identifying patients requiring urgent care, outperforming junior physicians whose sensitivity was 76.4 percent. The tool used natural language processing (NLP) to understand free-text symptom entries and integrated structured data from triage forms, including heart rate, oxygen saturation, and comorbidities. To avoid bias, physicians were blinded to the AI system’s predictions. Notably, in a focused analysis of chest pain cases, the AI showed an area under the curve of 0.91 in detecting heart attacks, significantly higher than clinician performance. It also reduced time to triage decision, with a median time of 3.7 minutes compared to 6.1 minutes for human providers. The model was trained on over 200,000 anonymized triage records, giving it exposure to a wide range of clinical scenarios. Researchers emphasize that the AI is not meant to replace physician judgment but rather act as a safety net in high-pressure settings. With further study, it could serve as a clinical decision support tool that helps optimize patient flow and reduce delays in critical care. Implementation efforts are currently underway to test integration into live ED workflows and assess its value in reducing missed high-risk cases.
Relevant reading: AI outperforms junior physicians in emergency triage study
©2025 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.
