Exploring the potential of artificial intelligence models for triage in the emergency department.

Postgrad Med

Department of Biostatistics, Faculty of Medicine, Ataturk University, Erzurum, Turkey.

Published: November 2024

Objective: To perform a comparative analysis of the three-level triage protocol conducted by triage nurses and emergency medicine doctors with the use of ChatGPT, Gemini, and Pi, which are recognized artificial intelligence (AI) models widely used in the daily life.

Materials And Methods: The study was prospectively conducted with patients presenting to the emergency department of a tertiary care hospital from 1 April 2024, to 7 April 2024. Among the patients who presented to the emergency department over this period, data pertaining to their primary complaints, arterial blood pressure values, heart rates, peripheral oxygen saturation values measured by pulse oximetry, body temperature values, age, and gender characteristics were analyzed. The triage categories determined by triage nurses, the abovementioned AI chatbots, and emergency medicine doctors were compared.

Results: The study included 500 patients, of whom 23.8% were categorized identically by all triage evaluators. Compared to the triage conducted by emergency medicine doctors, triage nurses overtriaged 6.4% of the patients and undertriaged 3.1% of the yellow-coded patients and 3.4% of the red-coded patients. Of the AI chatbots, ChatGPT exhibited the closest triage approximation to that of emergency medicine doctors; however, its undertriage rates were 26.5% for yellow-coded patients and 42.6% for red-coded patients.

Conclusion: The undertriage rates observed in AI models were considerably high. Hence, it does not yet seem appropriate to solely rely on the specified AI models for triage purposes in the emergency department.

Download full-text PDF

Source
http://dx.doi.org/10.1080/00325481.2024.2418806DOI Listing

Publication Analysis

Top Keywords

emergency department
16
emergency medicine
16
medicine doctors
16
triage nurses
12
triage
10
artificial intelligence
8
intelligence models
8
models triage
8
emergency
8
april 2024
8

Similar Publications

Background: Vancomycin, an antibiotic with activity against methicillin-resistant Staphylococcus aureus (MRSA), is frequently included in empiric treatment for community-acquired pneumonia (CAP) despite the fact that MRSA is rarely implicated in CAP. Conducting polymerase chain reaction (PCR) testing on nasal swabs to identify the presence of MRSA colonization has been proposed as an antimicrobial stewardship intervention to reduce the use of vancomycin. Observational studies have shown reductions in vancomycin use after implementation of MRSA colonization testing, and this approach has been adopted by CAP guidelines.

View Article and Find Full Text PDF

Background: Endodontic emergencies, often presented as acute pain or swelling, constitute a substantial challenge in dental practice. While effective management emphasizes prompt intervention, antibiotics are typically indicated only when systemic signs and symptoms are present. There is limited research exists on evaluating the knowledge and clinical approach of dental practitioners in managing endodontic emergencies from our region of the world.

View Article and Find Full Text PDF

CCL3 as a novel biomarker in the diagnosis of necrotizing enterocolitis.

BMC Pediatr

December 2024

Department of Clinical Laboratory, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Infection and Immunity, Children's Hospital of Chongqing Medical University, NO 136 Zhongshaner Road, Yuzhong Distrit, Chongqing, 400014, China.

Objectives: Neonatal necrotizing enterocolitis (NEC) is a common intestinal disease that threatens the lives of newborns and is characterized by ischemic necrosis of the small intestine and colon. As early diagnosis of NEC improves prognosis, the identification of new or complementary biomarkers is of great importance. In this study, we evaluate the diagnostic value of CCL3 in NEC and compare its effectiveness with other commonly used biomarkers, such as procalcitonin (PCT) and C-reactive protein (CRP).

View Article and Find Full Text PDF

Background: Multidose iodinated contrast media (ICM) injectors have shown promise in reducing ICM waste. This study aims to evaluate the impact of patient volume on ICM waste reduction in multidose injectors.

Methods: CT studies performed over one-year period with a multidose injector at our emergency CT unit.

View Article and Find Full Text PDF

Background: Effective staff-to-staff and patient-provider communication in the Emergency Department (ED) is essential for safe, quality care. Routine wearing of Personal-Protective-Equipment (PPE) has introduced new challenges to communication. We aimed to understand the perspectives of ED staff about communicating while wearing PPE, and to identify factors contributing to communication success, breakdown, and repair.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!