A5040288576- Artificial Intelligence in Healthcare and Education
- Radiomics and Machine Learning in Medical Imaging
- Radiology practices and education
- Topic Modeling
- Intracranial Aneurysms: Treatment and Complications
- Cerebrovascular and Carotid Artery Diseases
- COVID-19 diagnosis using AI
- Fetal and Pediatric Neurological Disorders
- AI in cancer detection
- Inflammatory Bowel Disease
- Radiation Dose and Imaging
- Machine Learning in Healthcare
- Vestibular and auditory disorders
- Advanced Neuroimaging Techniques and Applications
- Nutrition and Health in Aging
- Vascular Procedures and Complications
- Ovarian cancer diagnosis and treatment
- Autoimmune and Inflammatory Disorders
- Genetic and rare skin diseases.
- Cancer Diagnosis and Treatment
- Spinal Dysraphism and Malformations
- Abdominal vascular conditions and treatments
- Infectious Diseases and Tuberculosis
- Central Venous Catheters and Hemodialysis
- Computational and Text Analysis Methods
Kırıkkale University
2024-2025
Universidad CES
2024
Sağlık Bilimleri Üniversitesi
2024
Ministry of Health
2024
Ankara 29 Mayis State Hospital
2024
State Hospital
2024
Ankara Bilkent City Hospital
2023-2024
Bilkent University
2022-2024
SSK İhtisas Hastanesi
2024
Istanbul 29 Mayis University
2024
Purpose: To investigate and compare the diagnostic performance of ChatGPT 3.5, Google Bard, Microsoft Bing, two board-certified radiologists in thoracic radiology cases published by The Society Thoracic Radiology. Materials Methods: We collected 124 "Case Month" from Radiology website between March 2012 December 2023. Medical history imaging findings were input into Bing for diagnosis differential diagnosis. Two provided their diagnoses. Cases categorized anatomically (parenchyma, airways,...
This study examined the performance of four different multimodal Large Language Models (LLMs)—GPT4-V, GPT-4o, LLaVA, and Gemini 1.5 Flash—on multiple-choice visual neuroanatomy questions, comparing them to a radiologist an anatomist. The employed cross-sectional design evaluated responses 100 questions sourced from Radiopaedia website. accuracy was analyzed using McNemar test. According results, demonstrated highest with rate 90%, while anatomist achieved 67%. Among LLMs, GPT-4o performed...
OBJECTIVE Tobacco smoking is among the factors known to significantly augment risk of untreated intracranial aneurysm (IA) growth and rupture. Smoking appears have a variable effect on different endovascular treatment modalities. The impact safety, efficacy, outcomes Woven EndoBridge (WEB) device use for wide-neck IAs has not been evaluated. This study aimed investigate WEB devices by status. METHODS A retrospective multicenter analysis was conducted data patients from 36 sites worldwide...
Abstract While the Woven EndoBridge (WEB) device has transformed treatment of wide-neck intracranial aneurysms, incomplete occlusion remains a significant challenge requiring better understanding contributing factors. A retrospective analysis was conducted on multicenter data from patients who underwent WEB for aneurysms between January 2011 and December 2022. Using machine learning models, Cox regression, time-stratified analyses, we evaluated factors associated with persistent occlusion,...
Background Recent studies have highlighted the diagnostic performance of ChatGPT 3.5 and GPT-4 in a text-based format, demonstrating their radiological knowledge across different areas. Our objective is to investigate impact prompt engineering on diagnosing thoracic radiology cases, highlighting how complexity prompts influences model performance. Methodology We conducted retrospective cross-sectional study using 124 publicly available Case Month examples from Thoracic Society Radiology...
This study aimed to evaluate the performance of large language models (LLMs) and multimodal LLMs in interpreting Breast Imaging Reporting Data System (BI-RADS) categories providing clinical management recommendations for breast radiology text-based visual questions.
Limited research has examined the clinical consequences of sarcopenia and myosteatosis in Crohn's disease. This study aimed to determine prevalence, risk factors, effects on prognostic outcomes disease patients who underwent magnetic resonance enterography.This retrospective observational included 116 enterography between January 2015 August 2021. Skeletal muscle index was ratio cross-sectional area skeletal muscles at L3 vertebral level square neck imaging. Sarcopenia defined as <38.5...
ABSTRACT Purpose This study evaluated the diagnostic accuracy and differential diagnosis capabilities of 12 Large Language Models (LLMs), one cardiac radiologist, three general radiologists in radiology. The impact ChatGPT-4o assistance on radiologist performance was also investigated. Materials Methods We collected publicly available 80 “Cardiac Case Month’’ from Society Thoracic Radiology website. LLMs Radiologist-III were provided with text-based information, whereas other visually...
Purpose: To investigate and compare the diagnostic performance of 10 different large language models (LLMs) 2 board-certified general radiologists in thoracic radiology cases published by The Society Thoracic Radiology. Materials Methods: We collected publicly available 124 “Case Month” from Radiology website between March 2012 December 2023. Medical history imaging findings were input into LLMs for diagnosis differential diagnosis, while independently visually provided their assessments....
Background: The role of the cerebellum in motor function is well recognized. However, its higher nervous system activities such as cognition, emotion, endocrine, and autonomic less known. present study aims to show direct dento-amygdala projections using a biotinylated dextran amine (BDA) tracer rats 3-tesla (T) high-resolution diffusion tensor imaging (DTI)-based tractography humans. Materials Methods: BDA was pressure injected into dentate nucleus Wistar albino rats. Labeled cells axons...
Background The optimal endovascular approach for acutely ruptured wide‐neck intracranial aneurysms remains uncertain, and the use of stent‐assisted coiling or flow diversion is controversial due to antiplatelet therapy requirements potential risks. Various techniques have been developed address these challenges, including balloon‐assisted (BAC) intrasaccular flow‐disruption. Woven EndoBridge (WEB) an device that has shown a favorable efficacy safety profile with minimal rebleeding rates. We...