Yale Researchers Join NIH Bridge2AI Program

Yale School of Medicine has received funding from the National Institutes of Health (NIH) Common Fund’s Bridge2AI program to tackle challenging medical problems and accelerate discovery through the use of artificial intelligence (AI).

NIH is investing $130 million over four years to bring diverse teams together and bridge the gap between the biomedical and AI research communities. Yale researchers, including co-principal investigators Wade Schulz, MD, PhD, assistant professor of laboratory medicine and of biostatistics, and Samah Fodeh, PhD, assistant professor of emergency medicine and of biostatistics, have joined the initiative and are developing training modules that connect the use of AI to the biomedical field.

“AI is a really useful tool for predicting disease outcomes using real-life data coming from the health care system,” says Fodeh. “We’re looking forward to bridging AI with biomedicine and helping faculty and researchers who have expertise in health care learn how AI can be used to help them in their own work.”

"The Bridge2AI program is going to transform the way AI researchers can work with real biomedical data to build predictive models that will accelerate biomedical discoveries and people's health,” says Lucila Ohno-Machado, MD, PhD, MBA, distinguished professor of medicine (biomedical informatics) and a principal investigator for the Bridge Center at University of California San Diego (UCSD).

Starting in January 2023, Ohno-Machado will serve as deputy dean for biomedical informatics at Yale School of Medicine and will lead the school’s new Section for Biomedical Informatics and Data Science. She says both her current and future schools will contribute significantly. “The collaboration among Yale, UC San Diego, and many other institutions around the country will show how the multi-institutional team can truly achieve much more than any individual institution would ever be able to."

The use of AI to analyze complex datasets presents a groundbreaking opportunity to answer questions previously beyond the reach of biomedical researchers. The Bridge2AI program will support the generation of accessible, AI-ready datasets that can be used multiple times for a range of medical challenges.

“It’s very expensive to create these datasets for single projects,” says Schulz. “So being able to generate flagship datasets and make them broadly accessible for reuse is critical. I believe this project will stimulate research and development related to these machine-learning systems.”

Using these AI tools, however, requires a comprehensive understanding of health informatics. The Yale team will create training materials to foster the skills needed for successful machine learning analysis, including written or online lectures and mentorship programs, with a special focus on scientists from and who serve underrepresented communities.

“After the onset of COVID, we saw that underrepresented communities were hit the hardest,” says Fodeh. “One of our main goals is to ensure equity in education and reach those communities and facilitate their access to learning initiatives.”

Fodeh hopes their work will draw a diverse group of researchers to health informatics. Because different communities face their own unique challenges, reaching scientists from many backgrounds is essential for identifying and addressing a wide range of needs.

“I just love the overall focus of this inclusive program. I’m looking forward to serving underrepresented communities and women and improve the quality of mentoring we offer,” she says. “We hope to provide the best learning environment for these individuals.”

Research described in this article is supported by award number OT2OD032742 from the NIH Common fund.