(Toronto, September 21, 2026) Harnessing generative AI to help doctors make critical medical decisions requires more than just raw technology. A new review study in the Journal of Medical Internet Research published by JMIR Publications shows that adapting existing language models is the key to ensuring they can safely and effectively assist with clinical diagnoses, patient triaging, and treatment planning in real-world health care settings.
The research team, led by Anshum Patel, MD, and Joseph Y Cheung, MD, MS, analyzed 35 recent studies to understand how different customization methods affect AI performance and found that while standard large language models (LLMs) are capable, they must be tailored specifically for medical environments to be truly reliable. When models were connected directly to trusted medical databases or retrained on specific clinical guidelines, accuracy greatly improved, with some systems matching the diagnostic performance of human doctors.
The most successful approach was found to depend heavily on the specific medical task. For narrow, focused tasks like detecting cancer in medical images, retraining the AI on specific data worked best. For tasks that require reasoning through complex guidelines, linking the AI to live databases was highly effective. However, the researchers determined that the best performance came from hybrid systems, which combine both methods to manage complicated workflows like stroke triage and oncology cases.
“There is no single best way to adapt AI for health care. The right approach depends on the clinical task, and the next step is making sure these systems are safe, reliable, and useful in real-world patient care,” says Anshum Patel.
While these findings are promising, the researchers note that the vast majority of studies on these AI systems are based on past medical records rather than live patient testing. Before these advanced tools are widely adopted in hospitals, additional prospective, real-world testing is needed to guarantee patient safety and ensure the technology works reliably across different clinical environments.
Read the full study, titled “Fine-Tuning, Retrieval-Augmented Generation, and Hybrid Adaptation of Language Models for Clinical Decision-Making in Health Care: Systematic Review,” here.
Please cite as:
Patel A, Khand Y, Vallamchetla S, Li P, Tao C, Cheung J. Fine-Tuning, Retrieval-Augmented Generation, and Hybrid Adaptation of Language Models for Clinical Decision-Making in Health Care: Systematic Review. J Med Internet Res 2026;28:e104092
URL: https://www.jmir.org/2026/1/e104092
DOI: 10.2196/104092
About JMIR Publications
JMIR Publications is a leading open access publisher of digital health research and a champion of open science. With a focus on author advocacy and research amplification, JMIR Publications partners with researchers to advance their careers and maximize the impact of their work. As a technology organization with publishing at its core, we provide innovative tools and resources that go beyond traditional publishing, supporting researchers at every step of the dissemination process. Our portfolio features a range of peer-reviewed journals, including the renowned Journal of Medical Internet Research.
To learn more about JMIR Publications, please visit jmirpublications.com or connect with us via X, LinkedIn, YouTube, Facebook, Bluesky, and Instagram.
Head office: 130 Queens Quay East, Unit 1100, Toronto, ON, M5A 0P6 Canada
Media contact: communications@jmir.org
The content of this communication is licensed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, published by JMIR Publications, is properly cited.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.
Leave a Reply