Volume 12, Issue 2 (Summer 2025)
DSME 2025, 12(2): 215-224 |
Back to browse issues page
Research code: --
Ethics code: ---
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Deylami M, Raad H, Zarei T, Sayadinia M, Razavi B B M, Mehrpour S et al . Stroke GPT: a spoken artificial intelligence for the evidence review of stroke prevention and acute care. DSME 2025; 12 (2) :215-224
URL: http://dsme.hums.ac.ir/article-1-484-en.html
URL: http://dsme.hums.ac.ir/article-1-484-en.html
Mansour Deylami 
, Hanieh Raad , Tayyebeh Zarei , Mehrdad Sayadinia , Bibi Bibi Mona Razavi , Somayyeh Mehrpour , Maryam Ziaei
, Hanieh Raad , Tayyebeh Zarei , Mehrdad Sayadinia , Bibi Bibi Mona Razavi , Somayyeh Mehrpour , Maryam Ziaei
Associated Professor, Emergency Medicine, Zahedan University of Medical Sciences, Zahedan, Iran.
Abstract: (2057 Views)
Introduction: This study used large linguistic artificial intelligence models to create a user-friendly interface for reviewing high-level evidence on stroke prevention and acute care.
Information sources or data: A literature review of the research background was conducted between 2010-2024, focusing on general reviews of databases such as PubMed and Cochrane Library. The collected articles were used to design a chat bot using the meta-llama/Meta-Llama model with specific pre-processing to increase the accuracy and relevance of the answer.
Selection methods for study: Only guidelines and umbrella reviews that are systematic reviews of review studies were used in this database. Embedded on a website, the model uses data from more than 1,000 studies summarized in 228 systematic reviews, covering more than 1 million patients. The chatbot answers questions with evidence-based references and keeps the focus on the training data without wandering off into unrelated topics.
Conclusion: The results of this study demonstrated the ability of the model to provide detailed and sourced responses compared to the standard ChatGPT interface. But this model is only limited to English language.
Information sources or data: A literature review of the research background was conducted between 2010-2024, focusing on general reviews of databases such as PubMed and Cochrane Library. The collected articles were used to design a chat bot using the meta-llama/Meta-Llama model with specific pre-processing to increase the accuracy and relevance of the answer.
Selection methods for study: Only guidelines and umbrella reviews that are systematic reviews of review studies were used in this database. Embedded on a website, the model uses data from more than 1,000 studies summarized in 228 systematic reviews, covering more than 1 million patients. The chatbot answers questions with evidence-based references and keeps the focus on the training data without wandering off into unrelated topics.
Conclusion: The results of this study demonstrated the ability of the model to provide detailed and sourced responses compared to the standard ChatGPT interface. But this model is only limited to English language.
Type of Study: Review |
Subject:
Special
Received: 2025/02/18 | Accepted: 2025/09/14 | Published: 2025/06/22
Received: 2025/02/18 | Accepted: 2025/09/14 | Published: 2025/06/22
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC-By-NC), which permits use, distribution, and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.
Contact Information
© 2015 All Rights Reserved | Development Strategies in Medical Education
Designed & Developed by : Yektaweb
Development Strategies in Medical Education by http://dsme.hums.ac.ir/ is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.