Background: Advancements in ChatGPT are transforming medical education by providing new tools for assessment and learning, potentially enhancing evaluations for doctors and improving instructional effectiveness.
Objective: This study evaluates the performance and consistency of ChatGPT-3.5 Turbo and ChatGPT-4o mini in solving European Portuguese medical examination questions (2023 National Examination for Access to Specialized Training; Prova Nacional de Acesso à Formação Especializada [PNA]) and compares their performance to human candidates.
Methods: ChatGPT-3.5 Turbo was tested on the first part of the examination (74 questions) on July 18, 2024, and ChatGPT-4o mini on the second part (74 questions) on July 19, 2024. Each model generated an answer using its natural language processing capabilities. To test consistency, each model was asked, "Are you sure?" after providing an answer. Differences between the first and second responses of each model were analyzed using the McNemar test with continuity correction. A single-parameter t test compared the models' performance to human candidates. Frequencies and percentages were used for categorical variables, and means and CIs for numerical variables. Statistical significance was set at P<.05.
Results: ChatGPT-4o mini achieved an accuracy rate of 65% (48/74) on the 2023 PNA examination, surpassing ChatGPT-3.5 Turbo. ChatGPT-4o mini outperformed medical candidates, while ChatGPT-3.5 Turbo had a more moderate performance.
Conclusions: This study highlights the advancements and potential of ChatGPT models in medical education, emphasizing the need for careful implementation with teacher oversight and further research.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.2196/65108 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Midlife financial stress and cognitive and physical impairments in older age: The role of potentially modifying factors.
Soc Sci Med
January 2025
Division of Clinical Geriatrics, Center for Alzheimer's Research, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden; Ageing Epidemiology Research Unit (AGE), School of Public Health, Faculty of Medicine, Imperial College London, London, United Kingdom.
Background: Financial stress is an important source of chronic stress and has been associated with cognitive and physical impairments. The goal of this study was to investigate whether financial stress is associated with cognitive and physical impairment and their combination, the role of potential modifiable factors and potential sex differences.
Methods: The Cardiovascular Risk Factors, Aging, and Dementia population-based cohort study from Finland was used (n = 1497) (baseline data collected 1972-1987, mean age 50 years).
A survey of Swedish radiographer's need for knowledge at advanced level.
Radiography (Lond)
March 2025
Department of Health, Medicine and Caring Sciences, Linköping University, Sweden; Center for Medical Image Science and Visualization, Linköping University, Linköping, Sweden; Department of Radiology in Linköping, Sweden.
Introduction: There are uncertainties about whether current advanced-level courses provide the knowledge needed to develop the profession for radiographers in Sweden. The aim of this study was to investigate Swedish radiographers' perceived need for additional post-registration knowledge in their profession and their need for education at advanced level.
Methods: Swedish radiographers were invited to participate in a national electronic survey between November and December 2022.
Background: In Germany, the incidence of traumatic spinal cord injury is approximately 16 per million inhabitants per year. This article aims to present evidence-based diagnostic and therapeutic measures for the first 14 days after injury to minimize neural damage, prevent complications, and preserve functioning as much as possible.
Methods: After the formulation of key questions, systematic literature searches were carried out on multiple topics.
Enhancing Large Language Models with Retrieval-augmented Generation: A Radiology-specific Approach.
Radiol Artif Intell
March 2025
Department of Radiology & Biomedical Imaging, University of California, San Francisco (UCSF), San Francisco, Calif.
Retrieval-augmented generation (RAG) is a strategy to improve performance of large language models (LLMs) by providing the LLM with an updated corpus of knowledge that can be used for answer generation in real-time. RAG may improve LLM performance and clinical applicability in radiology by providing citable, up-to-date information without requiring model fine-tuning. In this retrospective study, a radiology-specific RAG was developed using a vector database of 3,689 articles published from January 1999 to December 2023.
View Article and Find Full Text PDFPerceived value and barriers of nursing specialty certifications among clinical nurses in Saudi Arabia: a cross-sectional study.
Front Med (Lausanne)
February 2025
College of Nursing, King Saud University, Riyadh, Saudi Arabia.
Introduction: Specialty nursing certifications reflect nurse's knowledge and competence in certain areas. Obtaining certification allows them to advance their careers and enhance patient care standards as their role and scope of responsibility expands. This study aimed to understand how nurses view specialty certification and related challenges in three university hospitals in Riyadh, Saudi Arabia.
View Article and Find Full Text PDFWant 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!