Achieving Equity in Generative Artificial Intelligence in Cardiovascular Medicine.

Area of Science:

• Cardiovascular medicine
• Artificial intelligence
• Health equity

Background:

• Generative artificial intelligence (AI) offers transformative potential in cardiovascular care.
• However, generative AI models risk exacerbating existing health inequities.
• Bias in representation, measurement, deployment, and contextual factors are key concerns.

Purpose of the Study:

• To review the risks of generative AI in perpetuating cardiovascular care inequities.
• To propose a framework for developing and implementing equitable generative AI.
• To explore the potential benefits of equitable generative AI in cardiovascular medicine.

Main Methods:

• Literature review of generative AI applications and biases in healthcare.
• Development of a framework for equitable generative AI based on design principles and operational strategies.
• Exploration of potential applications and benefits of generative AI in cardiovascular care.

Main Results:

• Generative AI models can perpetuate inequities through various biases (representation, measurement, deployment, contextual).
• A proposed framework emphasizes representation, contextualization, transparency, accountability, and community engagement for equitable AI.
• Strategies include digital equity and continuous recalibration of AI models.
• Potential benefits include improved access to care, enhanced patient-centeredness, better health literacy, and increased clinical trial participation.

Conclusions:

• Generative AI presents both risks and opportunities for advancing health equity in cardiovascular care.
• An equitable generative AI framework is crucial to mitigate bias and maximize benefits.
• Robust governance and oversight are recommended across the AI lifecycle.