University of Health Sciences Derince Training and Research Hospital Department of Emergency Medicine University of Health Sciences Derince Training and Research Hospital
Evaluating large language models for renal colic imaging recommendations: a comparative analysis of Gemini, copilot, and ChatGPT-4.0 Yavuz Yigit, Asım Enes Ozbek, Betul Dogru, Serkan Gunay, Baha AlKahlout International Journal of Emergency Medicine, 2025 Background The field of natural language processing (NLP) has evolved significantly since its inception in the 1950s, with large language models (LLMs) now playing a crucial role in addressing medical challenges. Objectives This study evaluates the alignment of three prominent LLMs—Gemini, Copilot, and ChatGPT-4.0—with expert consensus on imaging recommendations for acute flank pain. Methods A total of 29 clinical vignettes representing different combinations of age, sex, pregnancy status, likelihood of stone disease, and alternative diagnoses were posed to the three LLMs (Gemini, Copilot, and ChatGPT-4.0) between March and April 2024. Responses were compared to the consensus recommendations of a multispecialty panel. The primary outcome was the rate of LLM responses matching the majority consensus. Secondary outcomes included alignment with consensus-rated perfect (9/9) or excellent (8/9) responses and agreement with any of the nine panel members. Results Gemini aligned with the majority consensus in 65.5% of cases, compared to 41.4% for both Copilot and ChatGPT-4.0. In scenarios rated as perfect or excellent by the consensus, Gemini showed 69.5% agreement, significantly higher than Copilot and ChatGPT-4.0, both at 43.4% (p = 0.045 and < 0.001, respectively). Overall, Gemini demonstrated an agreement rate of 82.7% with any of the nine reviewers, indicating superior capability in addressing complex medical inquiries. Conclusion Gemini consistently outperformed Copilot and ChatGPT-4.0 in aligning with expert consensus, suggesting its potential as a reliable tool in clinical decision-making. Further research is needed to enhance the reliability and accuracy of LLMs and to address the ethical and legal challenges associated with their integration into healthcare systems.
AI-assisted decision-making in mild traumatic brain injury Yavuz Yigit, Mahmut Firat Kaynak, Baha Alkahlout, Shabbir Ahmed, Serkan Günay, Asim Enes Ozbek BMC Emergency Medicine, 2025 Objective This study evaluates the potential use of ChatGPT in aiding clinical decision-making for patients with mild traumatic brain injury (TBI) by assessing the quality of responses it generates for clinical care. Methods Seventeen mild TBI case scenarios were selected from PubMed Central, and each case was analyzed by GPT-4 (March 21, 2024, version) between April 11 and April 20, 2024. Responses were evaluated by four emergency medicine specialists, who rated the ease of understanding, scientific adequacy, and satisfaction with each response using a 7-point Likert scale. Evaluators were also asked to identify critical errors, defined as mistakes in clinical care or interpretation that could lead to morbidity or mortality. The readability of GPT-4’s responses was also assessed using the Flesch Reading Ease and Flesch-Kincaid Grade Level tools. Results There was no significant difference in the ease of understanding between responses with and without critical errors (p = 0.133). However, responses with critical errors significantly reduced satisfaction and scientific adequacy (p < 0.001). GPT-4 responses were significantly more difficult to read than the case descriptions (p < 0.001). Conclusion GPT-4 demonstrates potential utility in clinical decision-making for mild TBI management, offering scientifically appropriate and comprehensible responses. However, critical errors and readability issues limit its immediate implementation in emergency settings without oversight by experienced medical professionals.
Comparative analysis of emergency department admissions: A multi-center study on patient characteristics and mortality before and during the early phase of pandemic in Turkey Emre Şanci, Başak Bayram, Ahmet Naci Emecen, Asim Enes Özbek, Emrah Çelik, Onursal Varlikli, Şenol Ergüney, Alper Akin Gözübüyük, Hüseyin Cahit Halhalli Medicine United States, 2025 The effect of the global healthcare events makes its initial impact to the emergency departments. As the vanguards of healthcare system emergency departments had handled the surge while undergoing physical changes to manage ongoing crisis. Therefore, for the preparation of further global events, admission rates and emergency department utilization of the pandemic requires further investigation. The aim of this study was to analyze the admissions in a multi-center scale and compare ICD codes and crude mortality rates during the early COVID-19 pandemic period and compare this data to the previous year of the outbreak. This was a multi-center study of 10 hospitals; including 2 tertiary and 8 secondary healthcare institutions. This study comparatively analyzed ED admissions between during pandemic and pre-pandemic periods. A total of 8,38,502 admissions were included in the study (5,72,443 vs 2,66,059; pre pandemic period vs during pandemic period, respectively). There was a significant difference between age, gender, ED waiting time, triage color, and hospitalization rates (P < .001) between 2 periods. Comparison of admission international statistical classification of diseases and related health problems 10th revision codes were significantly different between pre-pandemic and pandemic periods for trauma, cardiovascular, neurology, and respiratory codes (P < .001, P < .001, P < .001, P = .024, respectively). The findings of this study suggest that a pandemic affects many mechanisms in the emergency healthcare systems. This influence might have affected the ED admission rates, hospital admission rates, and mortality rates.
Comparison of Standard Method and Triple Airway Maneuver on LMA Insertion Times in Prehospital Cardiac Arrest Simulation Ahmet Emir Sarı, Asım Enes Özbek, Emre Şancı, Hüseyin Cahit Halhallı, Merve Akın Özdemir, Akin Güllü, Erdem Aydin, Burhan Pehlivan, Hakan Özerol Prehospital Emergency Care, 2025 OBJECTIVES In this study, we hypothesized that in prehospital ambulance environments, the use of the triple airway maneuver-which facilitates airway patency in the mannequin-may allow for faster and easier placement of the Laryngeal Mask Airway (LMA). This study aimed to evaluate the effect of the triple airway maneuver on the LMA insertion times of paramedics with and without chest compression. METHODS This study was designed as a randomized, prospective, crossover simulated manikin study. Paramedics who were working in the Prehospital Command and Control Center were informed about the study. A randomized, crossover study design was used to reduce the learning curve of the participants. For each participant, four scenarios were created in a randomized order: 1) standard method with chest compressions, 2) triple airway maneuver with chest compressions, 3) standard method without chest compressions, and 4) triple airway maneuver without chest compressions. The study was carried out in a moving ambulance in an empty area. The standard method was defined as passive manual support of the head from below. For the triple airway maneuver, an emergency medicine specialist served as the second operator. The LMA application time was defined as the time until the LMA was held, and effective ventilation was provided. The primary outcome of the study was defined as the comparison of LMA insertion times of the standard method and triple airway maneuver for each method. RESULTS Twenty-eight participants were included in the study. When all scenarios were compared LMA insertion times did not differ between triple airway maneuver and standard method groups (p = 0.730). During chest compressions, no statistically significant difference was found between the LMA insertion times (mean difference: 0.57 seconds, 95% CI: -0.819 to 1.961; p = 0.406). Similarly, when chest compressions were not applied, no significant difference was observed between the groups (mean difference: 0.5001 seconds, 95% CI: -2.00 to 3.50; p = 0.675). CONCLUSIONS The findings of this study suggest that both the triple airway maneuver and the standard method may be effectively utilized in patients undergoing chest compressions as well as in those without chest compressions.
Inferior vena cava collapsibility index and trauma severity in elderly fall injuries Melike Erden, Huseyin Cahit Halhalli, Asim Enes Ozbek Revista Da Associacao Medica Brasileira, 2024 SUMMARY OBJECTIVE: Falls are a serious cause of morbidity and mortality among older people. One of the underlying causes of falls is dehydration. Therefore, ultrasonography has become an essential tool for evaluating volume status in the emergency department. However, the effect of volume status on falls in older people has not been evaluated before. The aim of this study was to determine the relationship between the inferior vena cava collapsibility index and the injury severity score in older patients who presented with fall-related injuries to the emergency department. METHODS: A total of 66 patients were included in the study. The injury severity score was used as the trauma severity score, and the Edmonton Frail Scale was used as the frailty scale. Volume status was evaluated with inferior vena cava collapsibility index. The primary outcome measure was defined as the correlation between inferior vena cava collapsibility index and injury severity score. Secondary outcome measures were defined as the effect of inferior vena cava collapsibility index and injury severity score on hospitalization and mortality. RESULTS: There was no significant correlation between injury severity score and inferior vena cava collapsibility index (p=0.342). Neither inferior vena cava collapsibility index nor injury severity score was an indicator of the mortality of these patients. However, injury severity score was an indicator of hospitalization. The mean Edmonton Frail Scale score was an indicator of mortality among older people who experienced falls (p=0.002). CONCLUSION: Inferior vena cava collapsibility index cannot be used to predict trauma severity in older patients who have experienced falls admitted to the emergency department.
Determination of the Effects of Change in Anxiety Level on Pain Perception in Patients who Present to Emergency Department due to Acute Pain: A Double Blind, Randomized, Controlled Trial Asım Enes ÖZBEK, Murat PEKDEMİR, Asim TOMO, Hüseyin ACAR, Ümit TURAL Anatolian Journal of Emergency Medicine, 2023 Aim: The aim of this study is to determine the level of pain and anxiety, and to investigate the effect of standard analgesic treatment and additional anxiolytic treatment on pain and anxiety in patients who presented to the emergency department due to acute pain.Material and Methods: This is a prospective, randomized, controlled, double-blind study. As the study group received dexketoprofen trometamol plus midazolam, the control group received dexketoprofen trometamol alone. At 0th, 30th, 60thand 120th minutes of treatment, patients' pain and anxiety levels were measured. Patients' overall anxiety levels were measured. The primary outcome measure was the comparison of pain and anxiety change at 0-30 minutes.Results: The study was conducted with 90 patients in each group. The median pain change was 33.5 (IQR, 38) for the control group and 30 (IQR, 33) for the study group, and the mean difference was 3.5 (95% CI; -7.2 to 14.2). The median anxiety change was 9.5 (IQR, 41) for the control group and 20 (IQR, 40) for the study group, and the mean difference was -10.5 (95% CI; -24.37 to 3.37). The rescue therapy needed, treatment satisfaction and preference to the same treatment in the future were similar between the control and the study group, respectively (26.7% vs 40%, p=0.058; 64% vs 57%, p=0.770; 90% vs 89%, p=0.802).Conclusion: In patients who present to the emergency department due to an acute pain complaint, adding anxiolytic treatment to the analgesic treatment does not contribute to a reduction of pain and anxiety.
The role of neutrophil-lymphocyte ratio and red blood cell distribution width in the classification of febrile seizures European Review for Medical and Pharmacological Sciences, 2017