Tell us about MedStar Health and your role there.
MedStar Health is a large healthcare system in the mid-Atlantic region, comprising 10 hospitals
and more than 300 care locations across Maryland, D.C., and Virginia. I serve as the Vice
President of Scientific Affairs for the MedStar Health Research Institute and direct the MedStar
Health National Center for Human Factors in Healthcare, leading a research team of about 30
people.
How did MedStar Health become involved with CHAI?
Our research has historically focused on patient safety and equity in digital health. As artificial
intelligence (AI) emerged as a disruptive technology in healthcare, we began engaging with
partners across the ecosystem. We discovered CHAI through connections with Duke University
and were excited by its mission to set standards and guidelines for promoting safety and equity
in AI.
What has been your experience participating in CHAI’s working groups?
My first major engagement was an in-person event at Stanford earlier this year, which brought
together partners from various sectors to share ideas about promoting safety and equity in AI.
Since then, I’ve been leading the predictive AI working group, where we’re developing
guidelines for the AI lifecycle. That event really laid out several ambitious goals for what we
need to do when it comes to promoting the safe and equitable use of AI.
How do you see CHAI’s work benefiting MedStar Health and other organizations?
Over the next 18 months, we hope to develop a set of guidelines applicable to the most typical
use cases in predictive and generative AI for healthcare facilities. These guidelines should be
developed with input from industry partners, state and federal agencies, patients, and
caregivers. The urgency is driven by the need to ensure safeguards are in place as healthcare
facilities are already adopting AI systems. The need is actually urgent; we have several
healthcare facilities that are adopting AI systems today that could really use this assistance.
What advice would you give to others considering joining CHAI’s working groups?
CHAI can play a key role in creating mutually agreed-upon certification criteria that promote
confidence in AI systems. We need both independent labs conducting evaluations based on
CHAI’s criteria and increased knowledge and capabilities for healthcare facilities to perform local
testing. CHAI can enable both certification processes and local testing capabilities.
Any final thoughts on your experience with CHAI?
Through CHAI events and working groups, I’ve connected with people from major tech
companies like Microsoft and Amazon Web Services, academic healthcare institutions like
Mayo Clinic, and smaller healthcare facilities that typically don’t have strong representation in
other groups. These networking opportunities are instrumental for informing guideline
development and future partnerships.
Media Inquiries: [email protected]