Yale researchers are building CARDIO to support AI cardiovascular health education during and after clinic visits. The team wants to reinforce what patients hear from clinicians. They also want patients to review guidance at their own pace.
The project is led by S. Raquel Ramos, an associate professor at the Yale School of Nursing. Ramos says she designed the tool to complement care teams. She points to gaps in time, staffing, and follow-up support.
Cardiovascular disease remains the leading cause of death in the United States. The American Heart Association says someone died from cardiovascular disease about every 34 seconds in 2023. Those numbers keep prevention high on the public health agenda.
What Yale Built and How It Works
Yale describes CARDIO as a purpose-built model, not a general chatbot. The team fine-tuned it on evidence-based cardiovascular guidance and patient education materials. They aim to deliver consistent explanations that match clinical standards.
A technical paper in the Journal of Medical Internet Research outlines the project’s development approach. The paper describes an interdisciplinary team across nursing, public health, and computer science. It also highlights careful data selection and iterative testing.
Yale says the team ran validation checks focused on accuracy and safety. They tested for guideline alignment and risky outputs. They also refined responses to improve clarity for patients. The project is moving toward a patient pilot.
How Patients May Use CARDIO After a Visit
In the planned pilot, clinicians will conduct a standard appointment and discuss risk factors. After that discussion, patients will use CARDIO for review and questions. The team wants to see how the tool fits into real care workflows.
Yale says patients can use suggested prompts or type their own questions. The prompts align with the American Heart Association’s Life’s Essential 8 framework. That framework covers behaviors and health factors tied to cardiovascular risk.
Life’s Essential 8 includes diet, physical activity, tobacco exposure, and sleep. It also includes blood pressure, cholesterol, blood sugar, and weight. Yale’s team expects this structure to keep answers focused and practical. They also hope it reduces confusion.
Benefits, Limits, and Guardrails in Health AI
Supporters of tools like CARDIO argue that better education can support earlier action. The CDC continues to describe cardiovascular disease as a major U.S. health burden. That burden drives interest in prevention and sustained patient engagement.
Researchers also warn about errors in AI medical outputs. False claims can mislead patients, especially in high-stakes contexts. Yale says it built CARDIO with safety testing and curated sources. The team also frames the tool as a supplement.
Ramos has described CARDIO as a way to extend education, not replace clinicians. The team wants to measure understanding, usability, and trust during the pilot. If results look strong, CARDIO could expand AI cardiovascular health education in routine care. It could also support follow-up conversations between visits.