Ayşenur Gür

Dr. Ayşenur Gür

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 April 2024 in Medicine
Verified
Journal Article
Open access
AGAyşenur Gür
Ayşenur Gür
YŞYeliz Şimşek
Yeliz Şimşek
The aim of the study is to determine the usefulness of base excess (BE) and creatine kinase (CK) in predicting the extent of damage to the extremities, the need for hemodialysis, and the likelihood of mortality in crush injuries. Our study included patients who were affected by the earthquakes that occurred in Kahramanmaras/Turkey on February 6, 2023 and were diagnosed with crush injuries. The study was a retrospective observational study. We used chi-square test, independent sample t test, analysis of variance (ANOVA) to examine whether CK and BE values can be used to predict damage to the extremities, hemodialysis requirement, and mortality. A total of 299 patients were included in the study. A statistically significant relationship was found between BE and extremity damage, hemodialysis requirement, and mortality (P < .005). A statistically significant difference was also seen in terms of extremity damage and hemodialysis requirement with CK (P < .001), while there was no statistically significant difference seen in mortality (P = .204). BE may serve as a predictive biomarker for the development of extremities damage, hemodialysis requirement, and mortality. CK is not predictive of mortality.

14 October 2025 in TURKISH JOURNAL OF MEDICAL SCIENCES
Verified
Journal Article
Open access
PDF file
IŞİsHak Şan
İsHak Şan
MÖMedi̇nE Akkan Öz
Medi̇nE Akkan Öz
Mehmet YortanlıMehmet Yortanlı
Mehmet Yortanlı
MGMurat Genç
Murat Genç
BBBensu Bulut
Bensu Bulut
MGMustafa Önder Gönen
Mustafa Önder Gönen
Background/aim: This study evaluated the accuracy rates and response consistency of four different large language models (ChatGPT- 4o, Gemini 2.0, Claude 3.5, and DeepSeek R1) in answering questions from the Emergency Medicine Fellowship Examination (YDUS), which was administered for the first time in Türkiye. Materials and methods: In this observational study, 60 multiple-choice questions from the Emergency Medicine YDUS administered on 15 December 2024, were classified as knowledge-based (n = 26), visual content (n = 2), and case-based (n = 32). Each question was presented three times to the four large language models. The models’ accuracy rates were evaluated according to overall accuracy, strict accuracy, and ideal accuracy criteria. Response consistency was measured using Fleiss’ Kappa test. Results: The ChatGPT-4o model was the most successful in terms of overall accuracy (90.0%), while DeepSeek R1 showed the lowest performance (76.7%). Claude 3.5 (83.3%) and Gemini 2.0 (80.0%) demonstrated moderate success. When analyzed by category, ChatGPT-4o achieved the highest success with 92.3% accuracy in knowledge-based questions and 90.6% in case-based questions. In terms of response consistency, the Claude 3.5 model (Fleiss’ Kappa = 0.68) showed the highest consistency, while Gemini 2.0 (Fleiss’ Kappa = 0.49) showed the lowest. Inconsistent hallucinations were more frequent in the Gemini 2.0 and DeepSeek R1 models, whereas persistent hallucinations were less common in the ChatGPT-4o and Claude 3.5 models. Conclusion: Large language models can achieve high accuracy rates for knowledge and clinical reasoning questions in emergency medicine but show differences in terms of response consistency and hallucination tendency. While these models have significant potential for use in medical education and as clinical decision support systems (CDSS), they need further development to provide reliable, up-to-date, and accurate information.

12 September 2025 in PLOS ONE
Verified
Journal Article
Open access
PDF file
BBBensu Bulut
Bensu Bulut
MÖMedine Akkan Öz
Medine Akkan Öz
MGMurat Genç
Murat Genç
AGAyşenur Gür
Ayşenur Gür
Mehmet YortanlıMehmet Yortanlı
Mehmet Yortanlı
ZUZekeriya Uykan
Zekeriya Uykan
Background: This study evaluates the diagnostic performance of three multimodal large language models (LLMs)—ChatGPT-4o, Gemini 2.0, and Claude 3.5—in identifying pneumothorax from chest radiographs. Methods: In this retrospective analysis, 172 pneumothorax cases (148 patients aged >12 years, 24 patients aged ≤12 years) with both chest radiographs and confirmatory thoracic CT were included from a tertiary emergency department. Patients were categorized by age and pneumothorax size (small/large). Each radiograph was presented to all three LLMs accompanied by basic symptoms (dyspnea or chest pain), with each model analyzing each image three times. Diagnostic accuracy was evaluated using overall accuracy (all three responses correct), strict accuracy (≥2 responses correct), and ideal accuracy (≥1 response correct), alongside response consistency assessment using Fleiss’ Kappa. Results: In patients older than 12 years, ChatGPT-4o demonstrated the highest overall accuracy (69.6%), followed by Claude 3.5 (64.9%) and Gemini 2.0 (57.4%). Performance was significantly poorer in pediatric patients across all models (20.8%, 12.5%, and 20.8%, respectively). For large pneumothorax in adults, ChatGPT-4o showed significantly higher accuracy compared to small pneumothorax (81.6% vs. 42.2%; p < 0.001). Regarding consistency, Gemini 2.0 demonstrated excellent reliability for large pneumothorax (Kappa = 1.00), while Claude 3.5 showed moderate consistency across both pneumothorax sizes. Conclusion: This study, the first to evaluate these three current multimodal LLMs in pneumothorax identification across different age groups, demonstrates promising results for potential clinical applications, particularly for adult patients with large pneumothorax. However, performance limitations in pediatric cases and with small pneumothoraces highlight the need for further validation before clinical implementation.

1 October 2025 in Turkish Journal of Trauma and Emergency Surgery
Verified
Journal Article
Open access
BBBensu Bulut
Bensu Bulut
Mehmet YortanlıMehmet Yortanlı
Mehmet Yortanlı
AGAyşenur Gür
Ayşenur Gür
MÖMedine Akkan Öz
Medine Akkan Öz
HMHüseyin Mutlu
Hüseyin Mutlu
Scaphoid fractures account for 60%-70% of wrist traumas, with delayed diagnosis leading to avascular necrosis and functional impairment. Traditional radiographic assessment remains challenging due to anatomical complexity and overlapping structures. This study evaluated three next-generation large language models (LLMs) (ChatGPT-4o, Gemini 2.0, and Claude 3.5) for their ability to detect scaphoid fractures and determine surgical indications. A retrospective observational study was conducted at Ankara Etlik City Hospital (October 2022 - January 2025) including 300 patients (150 with computed tomography confirmed (CT-confirmed) scaphoid fractures and 150 without fractures), aged 18-65 years, who presented to the emergency department (ED) with wrist trauma. Three-view wrist radiographs were presented to each LLM on three separate days. Diagnostic accuracy was assessed using overall accuracy (all three responses correct), strict accuracy (≥2 correct responses), and ideal accuracy (≥1 correct response). Response consistency was evaluated using Fleiss' kappa coefficient. Surgical indications were determined based on fracture displacement criteria. Claude 3.5 demonstrated superior sensitivity (57.1%) compared to Gemini 2.0 (18.2%) and ChatGPT-4o (9.1%) for fracture detection (p<0.001). Ideal accuracy rates were 79.3%, 36.0%, and 17.3%, respectively. Specificity remained uniformly low across models (43.1%-43.8%). All models performed better in non-fracture cases, with ideal accuracy exceeding 83%. Response consistency was moderate for all models (κ=0.36-0.41). For surgical indication assessment, Claude 3.5 identified 37.0% of cases requiring surgery, compared to ChatGPT-4o (34.1%) and Gemini 2.0 (24.4%), with correct determination rates of 73.7%, 71.4%, and 80.0%, respectively. Current LLMs demonstrate insufficient diagnostic accuracy for independent clinical use in scaphoid fracture detection. Claude 3.5's 57.1% sensitivity indicates that these technologies require substantial improvement before clinical deployment. However, their moderate performance in surgical decision-making suggests potential utility as assistive tools when combined with specialist expertise. Further development focusing on musculoskeletal-specific training is essential.

23 May 2026 in BMC Emergency Medicine
Verified
Journal Article
Open access
AGAyşenur Gür
Ayşenur Gür
RYRamiz Yazici
Ramiz Yazici
HMHuseyin Mutlu
Huseyin Mutlu
MÖMedine Akkan Öz
Medine Akkan Öz
YYYunus Yatmaz
Yunus Yatmaz
MKMustafa Sırrı Kotanoğlu
Mustafa Sırrı Kotanoğlu
Background: Crush syndrome is a life-threatening consequence of prolonged muscle compression, and acute kidney injury (AKI) is a major determinant of morbidity and mortality. Rapid AKI risk stratification at emergency department (ED) presentation remains challenging. We evaluated whether the lactate dehydrogenase-to-albumin ratio (LDAR) measured at ED admission could discriminate creatinine-defined AKI among patients with crush syndrome. Methods: We conducted a retrospective cohort study in the ED of a Level I trauma center from 1 January 2019 to 31 December 2025. Adults (≥ 18 years) meeting a predefined operational definition of crush syndrome, including traumatic compression injury, biochemical rhabdomyolysis, and compatible clinical or systemic features, were included. AKI was defined using serum creatinine-based Kidney Disease: Improving Global Outcomes criteria with a predefined approach for baseline creatinine determination. LDAR was calculated as admission lactate dehydrogenase divided by admission albumin. Discrimination was assessed using receiver operating characteristic (ROC) analysis with area under the curve (AUC) and 95% confidence intervals. Independent associations with AKI and RRT were examined using multivariable logistic regression. Results: Among 512 patients, 180 (35.2%) were classified as having creatinine-defined AKI; 374 (73.0%) were male. Hemodialysis was required in 78/180 (43.3%) patients with AKI, and in-hospital mortality was higher in patients with AKI (15.0% vs. 0.6%). Creatine kinase (CK) showed the highest discrimination for AKI (AUC 0.977, 95% CI 0.964–0.989), followed by LDAR (AUC 0.966, 95% CI 0.953–0.980). A Youden-derived LDAR cut-off of ≥ 23.59 yielded 86.7% sensitivity and 96.7% specificity. In addition, an LDAR cut-off of ≥ 22.68 predicted RRT requirement in patients with CS with 97.4% sensitivity and 77.6% specificity. Conclusions: LDAR measured at ED presentation showed high discrimination for AKI and RRT in crush syndrome. Although not a stand-alone treatment trigger, it may support early renal risk stratification when interpreted together with CK, renal function, electrolyte status, and clinical severity. Clinical trial number: Not applicable.

1 January 2024 in Emergency Medicine International
Verified
Journal Article
Open access
YŞYeliz Şimşek
Yeliz Şimşek
AGAyşenur Gür
Ayşenur Gür
Background: Physical examination and computed tomography angiography (CTA) are used for diagnosing arterial injury in extremity trauma. In recent years, CTA has been overused to obtain more objective data. Our study aimed to investigate the effect of using CTA for the management of patients with extremity penetrating injuries, specifically in cases where vascular injury was not detected during initial examination.Methods: This retrospective study included patients with penetrating trauma who underwent CTA of the extremities. The demographic data, mechanism of injury, the side of injury, initial vascular exam (normal, soft signs, and hard signs), radiological results, and any orthopedic and vascular intervention performed were recorded. The χ2 test was used for independent variables. A significance level of p < 0.05 was used. We compared the sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) for physical exam and CTA for identifying arterial injury requiring intervention.Results: Of the 252 patients included in the study, 29 (21.5%) had abnormal vascular physical examination while 26 (10.3%) had an abnormal CTA. The NPV of the hard sign for identifying vascular injury was 95.4%, while the sensitivity was 57.7%, specificity was 100%, and PPV was 100%. The NPV of routine physical examination to determine the requirement for vascular intervention was 100%. The sensitivity and PPV of the soft sign in determining the need for vascular intervention were 65.4% and 77.3%, respectively.Conclusion: Vascular injury was present in all cases that had positive hard signs. CTA imaging and vascular intervention are not necessary in patients who exhibit no hard and/or soft indicators during a thorough physical examination.