Large Language Models in Cardiology: Systematic Review
Clinical Snapshot
PICO Framework
| P — Population | Healthcare professionals and patients in cardiology settings |
| I — Intervention | Large language models (LLMs) including ChatGPT-3.5, ChatGPT-4, and Bing Chat |
| C — Comparator | Traditional methods, physician performance, or hospital materials |
| O — Outcomes | Accuracy, readability, performance across chronic conditions, acute scenarios, education, and diagnostic testing |
Bottom Line
This systematic review demonstrates that large language models show promise in cardiology applications, particularly for ECG interpretation and educational tasks, but significant limitations remain. ChatGPT-4 outperformed emergency physicians in ECG interpretation (91% vs 77% accuracy) and provided scientifically adequate patient education materials. However, critical gaps exist in emergency guidance, with Bing Chat omitting key myocardial infarction first aid steps in up to 45% of cases. Readability remains problematic, often requiring college-level comprehension when patient materials should target lower reading levels. The evidence base consists primarily of small simulation studies rather than real-world clinical validation. While LLMs may serve as useful adjuncts for education and diagnostic support, they cannot replace clinical judgment and require careful human oversight, particularly in acute care scenarios.
Key Findings
P Value: P < 0.001 for ECG interpretation comparison
Effect Size: ChatGPT-4 ECG interpretation: 91% accuracy vs 77% for emergency physicians
Primary Outcome: Variable LLM performance across cardiology tasks
Nnt Or Sensitivity: ChatGPT-3.5 heart failure questions: 91% accuracy (43/47)
Confidence Interval: Not consistently reported across studies
Clinical Application
Limited by readability issues (college-level comprehension required), incomplete emergency guidance, and need for human oversight Relevant for Australian cardiology practice, but TGA approval would be required for clinical decision support tools. RACGP guidelines on digital health tools would apply. No specific PBS implications as these are information tools rather than therapeutic interventions Cardiologists, emergency physicians, medical educators, and patients seeking cardiac health information
Abstract
BACKGROUND: Large language models (LLMs) are increasingly used in health care, but their role in cardiology has not yet been systematically evaluated. OBJECTIVE: This review aimed to assess the applications, performance, and limitations of LLMs across diverse cardiology tasks, including chronic and progressive conditions, acute events, education, and diagnostic testing. METHODS: A systematic search was conducted in PubMed and Scopus for studies published up to April 14, 2024, using keywords related to LLMs and cardiology. Studies evaluating LLM outputs in cardiology-related tasks were included. Data were extracted across 5 predefined domains and the risk of bias was assessed using an adapted QUADAS-2 tool (developed by Whiting et al at the University of Bristol). The review protocol was registered in PROSPERO (CRD42024556397). RESULTS: A total of 33 studies contributed quantitative outcome data to a descriptive synthesis. Across chronic conditions, ChatGPT-3.5 (OpenAI) answered 91% (43/47) heart failure questions accurately, although readability often required college-level comprehension. In acute scenarios, Bing Chat omitted key myocardial infarction first aid steps in 25% (5/20) to 45% (9/20) of cases, while cardiac arrest information was rated highly (mean 4.3/5, SD 0.7) but written above recommended reading levels. In physician education tasks, ChatGPT-4 (OpenAI) demonstrated higher accuracy than ChatGPT-3.5, improving from 38% (33/88) to 66% (58/88). In patient education studies, ChatGPT-4 provided scientifically adequate explanations (5.0-6.0/7) comparable to hospital materials but at higher reading levels (11th vs 7th grade). In diagnostic testing, ChatGPT-4 interpreted 91% (36/40) electrocardiogram vignettes correctly, significantly better than emergency physicians (31/40, 77%; P< .001), but showed lower performance in echocardiography. CONCLUSIONS: LLMs show meaningful potential in cardiology, especially for education and electrocardiogram interpretation, but performance varies across clinical tasks. Limitations in emergency guidance and readability, as well as small in silico study designs, highlight the need for multimodal models and prospective validation.
References
- 1.Gendler, M., Nadkarni, G. N., Sudri, K., Cohen-Shelly, M., Glicksberg, B. S., Efros, O., Soffer, S., & Klang, E. (2026). Large Language Models in Cardiology: Systematic Review. JMIR Cardio, 10(1), e76734. https://doi.org/10.2196/76734
Related Research
Bioelectrochemistry (Amsterdam, Netherlands)
From data to diagnosis: A comprehensive review of machine learning-driven wearable sensors in healthcare.
2 Aug 2026
Ultrasound in medicine & biology
A Systematic Review: The Application of Attention Mechanisms in Medical Ultrasound Image Processing
2 Aug 2026
Journal of the American Medical Informatics Association : JAMIA
Explainable machine learning in healthcare: methods, interpretation, and applications for clinical research
2 Aug 2026
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service