Dependability of Large Language Models in Cardiovascular Medicine: A Scoping Review
Authors' conclusion
Although some LLM-generated responses to cardiovascular-related questions exhibit acceptable levels of quality, significant drawbacks persist. These include verbosity, inaccuracies, occasional misinformation, inconsistent outputs to identical questions, bias, and poor reproducibility.
Study context
- Population
- Studies on development, implementation and assessment of large language models (LLMs) in cardiovascular medicine, published August 2020–February 2025
- Participants
- 25 publications
- Outcome
- Performance of LLMs in responding to cardiology-related questions
Methodological detailsCochrane methodology, PRISMA guidelines, 3 databases (PubMed, Web of Science, Embase)
- Cochrane methodology
- PRISMA guidelines
- 3 databases (PubMed, Web of Science, Embase)
Main finding
As reported in the abstract
- Assessment methodology varied considerably across studies
- LLMs showed potential utility in cardiovascular decision-making, myocarditis management, cardiac arrest diagnosis and treatment, and image differentiation
Original publication
Dependability of Large Language Models in Cardiovascular Medicine: A Scoping Review. J Cardiothorac Vasc Anesth. Published 22 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →