Accelerating interdisciplinary innovation in eHealth and artificial intelligence

Launched in 2021, e-HAIL is a strategic initiative between Michigan Medicine and the University of Michigan College of Engineering that aims to make U-M a premier hub for eHealth and Artificial Intelligence (AI) to improve health and healthcare using technology. A vibrant and diverse community of clinicians and methodologists, e-HAIL focuses on collaboration, grant development, and infrastructure to support a multi-disciplinary approach to innovation in eHealth and AI. The initiative provides opportunities for researchers to connect with others who are interested in using AI and machine learning methodologies to conduct high-impact research on critical healthcare applications.

The creation of e-HAIL resulted from a growing awareness at Michigan Medicine of the importance of AI in healthcare as well as the wealth of AI expertise in the College of Engineering. U-M’s AI lab, for example, is one of the oldest in the nation, dating back to the 1980s. The MM Office of Research worked with the CoE to establish the joint e-HAIL program, which marries eHealth with AI and has since become a model for making AI expertise relevant to particular fields. 

“The MM Office of Research held a Strategic Planning Retreat in 2021 with the department chairs to review and refresh the Great Minds, Greater Discoveries research strategic plan formulated in 2019, right before the COVID-19 pandemic. During the retreat, they were asked to identify the main areas, from their respective disciplines, that had potential for research advancements based on collaborations across the University, external partners, and communities in the coming years,” says J. Henrike Florusbosch, PhD, e-HAIL program manager.

e-Health was one of the five strategic areas identified under the “Collaboration” rubric. “Leadership had great insight in understanding that this would be an important area to pursue,” she says. “The problems that AI can help solve in healthcare can only be solved if they are collaborative efforts. A single kind of expert can no longer solve these types of problems. We want to enable technological breakthroughs in core AI methodologies that mitigate against health inequities and positively impact the health of individuals and populations within the healthcare system and beyond, both in the United States and globally.”

Leadership Model

e-HAIL employs a joint leadership model with co-lead faculty conveners from MM and the CoE who work together with Florusbosch, including: Rada Mihalcea, Ph.D., Janice M. Jenkins Collegiate Professor of Computer Science and Engineering, Professor of Electrical Engineering and Computer Science, College of Engineering; and Akbar K. Waljee, M.D., M.Sc., AGAF, Assistant Dean, Global Health Research, Lyle C. Roll Endowed Professor, Professor, Learning Health Sciences, Professor, Internal Medicine, Michigan Medicine. 

The initiative also draws upon the insights of advisory conveners representing schools and colleges across campus, including the U-M Medical School, College of Engineering, and the College of Literature, Science, and the Arts.  “This is a nod to the fact that we have many connections and support faculty from all the schools and colleges,” says Florusbosch. e-HAIL members also come from the School of Information, School of Public Health, School of Nursing, the College of Pharmacy, and the Institute for Social Research.

Healthcare Applications

Florusbosch explains that researchers across the university are working on the development of AI tools and technologies for a variety of health research and biomedical domains. For example, work in surgical education uses computer vision to assist in training physicians. Natural Language Processing (NLP) experts work on health domains such as medical chatbots that can address health misinformation, provide bilingual health assistance, and train mental health providers to be better at delivering care.

“The funding landscape is evolving together with the growing interest in translating AI for better health outcomes and to enable advances in biomedicine,” she says. “For example, researchers used to be able to get funding for grants when they were developing a predictive model for conditions X, Y, and Z. But the bar is continuously being raised. Now, funders want to see that researchers can implement and evaluate the models in the real world. Working from the start with multidisciplinary teams provides an advantage and a potential area where we will see growth.” 

Bi-Weekly Meetings

Community building and collaboration are critical to eHAIL’s mission, and the initiative accomplishes this in several ways. One is with biweekly feedback and brainstorming sessions open to e-HAIL’S 100+ members only. “We are looking to help researchers who are ready to submit a new idea for external funding within the next few years and who have an active research agenda in this space,” says Florusbosch

The sessions are interactive and constructive and frequently center around a set of potential Specific Aims. “A research team presents their ideas and addresses specific aims they might want to propose. From there, the group works through the ideas and offers improvements,” she says. “The process creates an engaged community of AI and health experts that know and trust each other. So, when new grant opportunities come up, there are already established relationships where people have a sense of what others are doing in terms of work. These long-term collaborations where people see a mutual interest are essential.”

Funding Opportunities

As a resource, e-HAIL maintains a current list of funding opportunities in the AI and health space on its website, providing specifics about the purpose of the funding opportunity. These summaries are accessible to U-M faculty researchers across all schools and colleges interested in developing AI and machine learning to improve healthcare and are a great way to start understanding the funding landscape. Additional funding opportunities, including student summer support and data set creation awards, are available to e-HAIL members.

Global Impact

Florusbosch says combining core AI expertise with a particular content area is a model that works well. 

“The intersection of AI and eHealth is not restricted to the United States or countries in the global north. It's also very much of interest to countries in the global south or in lower resource countries,” she says. “This is something that Rada Mihalcea Akbar K. Waljee, and I, feel very strongly about: how AI for healthcare can have impacts globally. We're actively seeking out projects that extend beyond the borders of the United States or involve low-resource settings. We think this initiative has a lot of opportunity to have a global impact.”