Assessing the precision of artificial intelligence in ED triage decisions: Insights from a study with ChatGPT.

BACKGROUND: The rise in emergency department presentations globally poses challenges for efficient patient management. To address this, various strategies aim to expedite patient management. Artificial intelligence's (AI) consistent performance and rapid data interpretation extend its healthcare applications, especially in emergencies. The introduction of a robust AI tool like ChatGPT, based on GPT-4 developed by OpenAI, can benefit patients and healthcare professionals by improving the speed and accuracy of resource allocation. This study examines ChatGPT's capability to predict triage outcomes based on local emergency department rules. METHODS: This study is a single-center prospective observational study. The study population consists of all patients who presented to the emergency department with any symptoms and agreed to participate. The study was conducted on three non-consecutive days for a total of 72 h. Patients' chief complaints, vital parameters, medical history and the area to which they were directed by the triage team in the emergency department were recorded. Concurrently, an emergency medicine physician inputted the same data into previously trained GPT-4, according to local rules. According to this data, the triage decisions made by GPT-4 were recorded. In the same process, an emergency medicine specialist determined where the patient should be directed based on the data collected, and this decision was considered the gold standard. Accuracy rates and reliability for directing patients to specific areas by the triage team and GPT-4 were evaluated using Cohen's kappa test. Furthermore, the accuracy of the patient triage process performed by the triage team and GPT-4 was assessed by receiver operating characteristic (ROC) analysis. Statistical analysis considered a value of p < 0.05 as significant. RESULTS: The study was carried out on 758 patients. Among the participants, 416 (54.9%) were male and 342 (45.1%) were female. Evaluating the primary endpoints of our study - the agreement between the decisions of the triage team, GPT-4 decisions in emergency department triage, and the gold standard - we observed almost perfect agreement both between the triage team and the gold standard and between GPT-4 and the gold standard (Cohen's Kappa 0.893 and 0.899, respectively; p < 0.001 for each). CONCLUSION: Our findings suggest GPT-4 possess outstanding predictive skills in triaging patients in an emergency setting. GPT-4 can serve as an effective tool to support the triage process.

Diagnostic accuracy of apple watch ECG outputs in identifying dysrhythmias: A comparison with 12-Lead ECG in emergency department.

BACKGROUND: Wearable devices, particularly smartwatches like the Apple Watch (AW), can record important cardiac information, such as single­lead electrocardiograms (ECGs). Although they are increasingly used to detect conditions such as atrial fibrillation (AF), research on their effectiveness in detecting a wider range of dysrhythmias and abnormal ECG findings remains limited. The primary objective of this study is to evaluate the accuracy of the AW in detecting various cardiac rhythms by comparing it with standard ECG's lead-I. METHODS: This single-center prospective observational study was conducted in a tertiary care emergency department (ED) between 1.10.2023 and 31.10.2023. The study population consisted of all patients assessed in the critical care areas of the ED, all of whom underwent standard 12­lead ECGs for various clinical reasons. Participants in the study were included consecutively. An AW was attached to patients' wrists and an ECG lead-I printout was obtained. Heart rate, rhythm and abnormal findings were evaluated and compared with the lead-I of standard ECG. Two emergency medicine specialists performed the ECG evaluations. Rhythms were categorized as normal sinus rhythm and abnormal rhythms, while ECG findings were categorized as the presence or absence of abnormal findings. AW and 12­lead ECG outputs were compared using the McNemar test. Predictive performance analyses were also performed for subgroups. Bland-Altman analysis using absolute mean differences and concordance correlation coefficients was used to assess the level of heart rate agreement between devices. RESULTS: The study was carried out on 721 patients. When analyzing ECG rhythms and abnormal findings in lead-I, the effectiveness of AW in distinguishing between normal and abnormal rhythms was similar to standard ECGs (p = 0.52). However, there was a significant difference between AW and standard ECGs in identifying abnormal findings in lead-I (p < 0.05). Using Bland-Altman analysis for heart rate assessment, the absolute mean difference for heart rate was 0.81 ± 6.12 bpm (r = 0.94). There was strong agreement in 658 out of 700 (94%) heart rate measurements. CONCLUSION: Our study indicates that the AW has the potential to detect cardiac rhythms beyond AF. ECG tracings obtained from the AW may help evaluate cardiac rhythms prior to the patient's arrival in the ED. However, further research with a larger patient cohort is essential, especially for specific diagnoses.