Vanderbilt, Duke Awarded $1.25M to Improve Health Systems’ AI Oversight

Researchers from Vanderbilt University Medical Center (VUMC) and Duke University School of Medicine have been awarded a $1.25 million grant from the Gordon and Betty Moore Foundation to establish a framework to improve oversight of health systems’ artificial intelligence (AI) tools.

The project, titled “Measuring Artificial Intelligence (AI) Maturity in Healthcare Organizations,” will bring together experts from Duke, VUMC, the University of Iowa, and the Coalition for Health AI (CHAI) to outline the capabilities health systems need to establish in order to utilize trustworthy AI.

“The promise of AI for improving health and health care is great, but there is currently a wide gap between promise and reality,” said Peter Embí, MD, MS, professor and chair of the Department of Biomedical Informatics and senior vice president for Research and Innovation at VUMC, in the press release. “This work will produce new tools and capabilities that our health system needs to ensure that we select, deploy and monitor health AI to make health care more safe, effective, ethical and equitable for all.”

The researchers noted that health systems are already developing and deploying AI algorithms, but these efforts are currently being undertaken with limited oversight and gaps in both organizational infrastructure and resources.

These challenges make it difficult to document and track which algorithms are being deployed, who is accountable for them, who is monitoring them, and what the value of each tool is. This has led to concerns about the quality, safety, and fairness of health AI.

“If we are to realize the full potential of AI technologies, health systems must develop a more mature process for implementing these tools. Improving oversight of AI technology in health care systems is crucial for ensuring the safety and efficacy of patient care,” explained Michael Pencina, PhD, chief data scientist at Duke Health and the Duke School of Medicine’s vice dean for data science.

“With this grant, we’re taking a significant step toward ensuring safe, accountable, standards-based deployment of AI in health systems,” he continued.

The press release indicates that over the next year, the research teams will engage stakeholders from other health systems and organizations like CHAI to determine what strategies and infrastructure components are necessary for the implementation of trustworthy health AI.

“Creating a maturity model framework for health AI will enable health systems to identify their strengths and weaknesses when procuring and deploying AI solutions, ultimately driving the transformation of health care for the better,” said Nicoleta Economou, PhD, director of Algorithm-Based Clinical Decision Support Oversight at Duke AI Health.

This work underscores a broader push toward ensuring that healthcare AI is safe, effective, and equitable.

Later this week, Emory University will establish the Emory Empathetic AI for Health Institute (Emory AI.Health), which aims to bolster the development of AI tools to improve health outcomes for patients with a variety of conditions, including diabetes, cancer, and heart disease.

The institute is set to bring together experts from Emory, the Georgia Institute of Technology (Georgia Tech), the Atlanta Veterans Affairs (VA) Medical Center, and others to advance the development of cost-effective, equitable, and accessible healthcare AI algorithms.