Search Articles

View query in Help articles search

Search Results (1 to 10 of 371 Results)

Download search results: CSV END BibTex RIS

CSV download: Download all 371 search results (up to 5,000 articles maximum)

Perceptions, Usage, and Educational Impact of ChatGPT Among Medical Students in Germany: Cross-Sectional Mixed Methods Survey

Perceptions, Usage, and Educational Impact of ChatGPT Among Medical Students in Germany: Cross-Sectional Mixed Methods Survey

The following open-ended questions were included: “What concerns do you have when using Chat GPT?” “In which areas do you see potential in using Chat GPT?” “For which subjects or areas do you primarily use Chat GPT and why?” The survey was pretested for usability and technical functionality by 3 medical students who were not part of the final sample. Their feedback was used to refine the wording, layout, and navigation of the questionnaire before launching the final version via So Sci Survey.

Anna Fußhöller, Fabian Lechner, Nadine Schlicker, Felix Muehlensiepen, Andreas Mayr, Sebastian Kuhn, Martin Christian Hirsch, Johannes Knitza

JMIR Form Res 2025;9:e81484


Quality Assessment of Large Language Model–Generated Medical Dialogue for Clinical Vignettes: Evaluation Study

Quality Assessment of Large Language Model–Generated Medical Dialogue for Clinical Vignettes: Evaluation Study

Tools such as Chat GPT, developed by Open AI, possess advanced conversational capabilities and broad applicability [1], with an increasing number of applications in the medical field. From the perspective of conversational capabilities, generative AI–based chatbots can provide rehabilitation guidance and mental health support to patients and assist health care professionals in patient management [2]. Generative AI is versatile enough to have passed the Japanese National Medical Licensing Examination [3].

Yasutaka Yanagita, Daiki Yokokawa, Shiichi Ihara, Ryo Yoshida, Yoshihide Okano, Takanori Uehara

JMIR Form Res 2025;9:e80752


Applications, Challenges, and Prospects of Generative Artificial Intelligence Empowering Medical Education: Scoping Review

Applications, Challenges, and Prospects of Generative Artificial Intelligence Empowering Medical Education: Scoping Review

Another study comparing automatic scoring systems (Chat GPT-3.5 and Chat GPT-4) with manual scoring for article quality assessment found no significant difference between GPT-4-based scoring and human grading. This demonstrates the considerable potential of GAI to enhance the quality evaluation of articles [28].

Yuhang Lin, Zhiheng Luo, Zicheng Ye, Nuoxi Zhong, Lijian Zhao, Long Zhang, Xiaolan Li, Zetao Chen, Yijia Chen

JMIR Med Educ 2025;11:e71125


AI’s Accuracy in Extracting Learning Experiences From Clinical Practice Logs: Observational Study

AI’s Accuracy in Extracting Learning Experiences From Clinical Practice Logs: Observational Study

Research using Chat GPT, an LLM, has shown that it can apply codes to interview texts using a codebook, suggesting its potential for extracting competency-based evaluations from student descriptions [13]. However, owing to a lack of such research, aggregation accuracy remains uncertain. Determining the extent to which LLMs can aggregate items related to curriculum goals from learner descriptions may open up opportunities to leverage LLMs to monitor learner progress and enhance education quality.

Takeshi Kondo, Hiroshi Nishigori

JMIR Med Educ 2025;11:e68697


Author’s Response to Peer Reviews of “Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study”

Author’s Response to Peer Reviews of “Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study”

I use the latest version of Chat GPT (GPT-4). I have been using it for a while for my manuscripts, both in English and Portuguese, and it does an excellent job. As I am applying for a full article processing charge waiver and cannot afford professional copyediting services, I have leveraged advanced language models to assist with improving the manuscript, and it would be possible to do this again.

Jorge Guerra Pires

JMIRx Med 2025;6:e83417


Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study

Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study

Chat GPT has been extensively explored in bioinformatics since its release, as have LLMs in general. Even so, in bioinformatics, the traditional paradigm is to build a model with no concern as to how to integrate those models into something more user-friendly, such as a chatbot. The research in this field generally tends to be an exploration of the LLM as a language model only [7-9]. Studies tend to focus on what is called a chat-oriented conversational AI [4].

Jorge Guerra Pires

JMIRx Med 2025;6:e56090


ChatGPT-Based Chatbot for Help Quitting Smoking via Text Messaging: An Interventional Study

ChatGPT-Based Chatbot for Help Quitting Smoking via Text Messaging: An Interventional Study

LLM chatbots such as Open AI’s Chat GPT, Google’s Gemini, and Meta’s Llama allow for open-text queries and provide dynamic, natural language responses that are sensitive to context, nuance, and history and resemble a human conversation [14-16]. In 2025, a total of 34% of American adults and 58% of adults under 30 years of age in the United States, reported that they had used Chat GPT, with use rising from the previous year [17].

Lorien C Abroms, Christina N Wysota, Artin Yousefi, Tien-Chin Wu, David A Broniatowski

JMIR Form Res 2025;9:e79402


Performance of Large Language Models in Diagnosing Rare Hematologic Diseases and the Impact of Their Diagnostic Outputs on Physicians: Combined Retrospective and Prospective Study

Performance of Large Language Models in Diagnosing Rare Hematologic Diseases and the Impact of Their Diagnostic Outputs on Physicians: Combined Retrospective and Prospective Study

The top-10 accuracy (score range 0‐1) results indicated that Chat GPT-o1-preview performed best (0.703), successfully including the correct diagnosis in 70.3% of cases when generating 10 differential diagnoses. It was followed by Deep Seek-R1, Gemini Experimental 1206, Claude 3.5 Sonnet, Chat GPT-4o, Qwen-Max-2025-01-25, and Doubao-1.5-Pro-256k.

Hongbin Yu, Tian Chen, Xin Zhang, Yunfan Yang, Qinyu Liu, Chenlu Yang, Kai Shen, He Li, Wenjiao Tang, Xushu Zhong, Xiao Shuai, Xinmei Yu, Yi Liao, Chiyi Wang, Huanling Zhu, Yu Wu

J Med Internet Res 2025;27:e77334