Collaborating with AI to improve public health in Michigan

The Partnership for Public Service AI Center for Government™ is publishing a series of blogs to celebrate how AI and intelligent automation are being used in government to serve the public.    

Recently, we had the opportunity to speak with Joe Coyle, director of the Bureau of Infectious Disease Prevention and Deputy State Epidemiologist at the Michigan Department of Health and Human Services, to hear how machine learning and intelligent automation are supporting public health outcomes in the state of Michigan. 

_{Responses have been lightly edited for clarity.}

How are you thinking about or engaging AI in your work? 

“We’re thinking about theoretical uses and ways to modernize our workflows and our business to create automations and efficiencies so that we can utilize humans to do things that only humans can do. 

That is a huge component of infectious disease response and investigation: you need a human brain, you need human intuition, you need partnerships, you need to be able to communicate and have conversations with people.   

We want to be able to leverage and use large amounts of data so we can do the best public health response possible. But [public health professionals] don’t need to be deduplicating records or manually collecting or inputting information into spreadsheets. There’s potential for machine learning to automate rote tasks to allow us to do much more innovative work and focus time on things that are really the most impactful for keeping people healthy.  

Our job is to prevent people from getting sick and dying unnecessarily from things that are preventable, and I would much rather spend time having a conversation with somebody about how they could prevent that thing from being transmitted in their household, or to better understand what their barriers are to getting tested or treated, and helping enable them to have the opportunity to get treatment by linking them to other types of social services.  

If we can do better with electronic lab reports, electronic case reports and data exchange, then we have a better opportunity to really do what the public deserves and expects their governmental public health officials to be doing.” 

We appreciate this focus on the human! 

“Yes, and we work together. We’re fortunate that we have good communities of practice across the country and we learn a lot from our colleagues. One of the things I’m really excited about is the CDC Center for Forecasting and Outbreak Analytics’ Insight Net [a national outbreak analytics and disease modeling network], one of which is the University of Michigan School of Public Health, who we’ve worked with very closely.” 

Please tell us more about your work with the CDC CFA Insight Net 

“The real emphasis [of the CDC-funded centers] is to create value for governmental public health so we have a really close relationship with the University of Michigan School of Public Health. We meet with them on a weekly basis and it’s been very interesting for both of us to see each other’s way of operating and doing business and to share the considerations around things like AI large language models and what opportunities might be applicable to the work that we do. I’m excited to see what we’re able to create and stand up.” 

Is there an AI example that comes to mind? 

“One that’s been a topic of conversation for a long time is detecting aberrations in data. For example, I’ve run into more Shigella [a type of bacteria] cases than I normally would and I wonder if something is going on. But oftentimes, especially when you’re dealing with big data, you can’t really see the difference between numbers or the difference might be missed by a human eye. There are opportunities for [AI] models to really be able to flag something out of sync and then you can get some human eyes on it and figure out what the differences are and whether it’s above normal or below normal.  

That’s one of the things that we’re working on—figuring out how we could use AI to better identify certain aberrations in data that need to be looked into further, especially with infectious diseases, and better notice how data is clustering in space and time. There’s only so much that a human can do to monitor that comprehensively across the entire ecosystem and landscape of things that we do. 

We’re learning all the time what different opportunities might be and [we] work with our colleagues in government to see how we might be able to leverage some of those technologies to deliver better outcomes to communities.” 

---

Continuing the conversation  

The AI Center for Government champions AI innovators across all levels of government. If your agency is taking steps to lead AI well, we’d love to hear from you. Join us as we highlight real-world AI use cases and convene public-sector leaders from across the country to share tools and insights to lead confidently in the age of AI.  

We’re here to help!   

• Sign up for our newsletter. 
• Check out our AI Center for Government programs, resources and events.  
• Get in touch! Email us at [email protected].