HIMSSCast HIMSSCast: AI assurance labs could improve health equity
5 snips
Jan 24, 2025 Brigham Hyde, CEO of Atropos Health, delves into the transformative potential of AI assurance labs in healthcare. He advocates for standardization to enhance equity and transparency in AI applications. The conversation navigates the complexities of integrating multiple AI models and the necessary evaluations to prioritize patient care. Hyde addresses challenges in balancing AI assurance costs, especially for rural providers, and calls for collaboration among tech companies and healthcare institutions to overcome biases and promote fair access to innovative tools.
AI Snips
Chapters
Transcript
Episode notes
AI Vetting Standards
- Establish clear AI vetting standards to guide health systems and startups.
- Federal guidance, like the FDA 21st Century Cures Act, offers a good starting point.
AI Evaluation Pillars
- Evaluate AI models based on accuracy, bias, impact, and real-world consequences.
- Consider the stakes of AI actions, from simple alerts to treatment decisions.
AI and Equity
- AI assurance labs can report on equity, not ensure it, by detecting existing biases in data.
- AI tools offer the potential to address healthcare inequities by generating new evidence for diverse populations.