Transforming Cardio-Oncology Care Through AI-Driven Large Language Model Systems: A Roadmap for Future Implementation.

Summary

Customizing large language models with institutional data and real-time clinical guidelines provides a scalable framework for managing cardiovascular risks in cancer patients. This approach improves clinical decision support and patient triage by integrating AI assistants directly into electronic health record workflows.

Key result

GPT-4 demonstrated superior performance among evaluated models, including Gemini 1.5 and LLaMA-3-70B, for clinical application in cardio-oncology.