GUIDE-AI
Guidance for the Use, Implementation, Development, and Evaluation of Healthcare AI
Our Mission
Building and deploying healthcare AI solutions
Establishing evaluation and monitoring methods to assess deployed AI tools
Disseminating actionable learnings to support high-value AI-augmented care for all
Resources
GUIDE-AI and collaborators at Stanford Medicine Technology & Digital Solutions and HEAL-AI have developed helpful primers and starter kits for those interested in exploring responsible AI development and deployment at their organizations. See our Resources page for more information.
Areas of Focus
Our team supports the responsible use, evaluation, and monitoring of AI tools currently being used at Stanford Health Care to improve clinical workflows.
Implementing AI tools in a rapidly evolving landscape is an ongoing challenge. GUIDE-AI has helped develop methods, frameworks, and focused briefs to support best practices in health AI governance, ethics, and monitoring.
See the implementation tools that GUIDE-AI has developed in the Primers & Resources section of Our Work.
As the prevalence of AI in healthcare skyrockets, it's critical to understand what is worth building.
GUIDE-AI leads in this work by bringing together diverse innovators – including clinicians, data scientists, and other subject matter experts – to explore key problems to solve and accelerate from ideation to implementation.
How do we know if AI-augmented care is working?
GUIDE-AI has developed methods to evaluate AI models, and ensure that they are delivering results users can trust. From frameworks to assess LLM performance for medical tasks to evaluations of AI-mediated workflows intended to identify and screen patients at risk of specific conditions, GUIDE-AI helps provide assurance that models are performing as expected.