Related Experiment Video
Updated: Jan 17, 2026
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
1.0K
Comparative performance of large language models on cardiovascular certification simulation exam
Eesha Nachnani1, Kashish Goel2, Alexander E Sullivan2
1University School of Nashville, Nashville, TN.
American Heart Journal
|January 14, 2026
Summary
Artificial intelligence (AI) is increasingly used in medicine. A study found that only ChatGPT-4.0, among tested AI models, performed comparably to humans on a cardiovascular medicine exam.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Cardiovascular Medicine
Background:
- Artificial intelligence (AI) demonstrates growing capabilities in medical practice.
- AI tools are being evaluated for their potential to assist in clinical decision-making and examinations.
Purpose of the Study:
- To assess the performance of leading large-language models (LLMs) on a cardiovascular medicine board-style examination.
- To compare the efficacy of different AI platforms in a specialized medical domain.
Main Methods:
- Three popular LLMs (ChatGPT-4.0, Gemini, Bing AI) were evaluated.
- Performance was measured against a cardiovascular medicine board-style exam, with scores compared to human participants.
Main Results:
- ChatGPT-4.0 achieved a score comparable to human participants on the cardiovascular exam.
- Gemini and Bing AI did not reach similar performance levels in this specialized medical assessment.
Conclusions:
- ChatGPT-4.0 shows significant potential for medical knowledge assessment in cardiology.
- Further research is needed to evaluate the broader clinical applicability and limitations of AI in specialized medical fields.
Related Concept Videos
02:07Improving Translational Accuracy
3.5K
3.5K
02:07Improving Translational Accuracy
14.1K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
14.1K