Digital epidemiology and artificial intelligence have the potential to transform the public health model that has been in place for nearly a century. While the COVID-19 pandemic raised the profile of public health officers, AI and digital epidemiology can take the science to new heights. So, the time was perfect for the University of Arizona Mel and Enid Zuckerman College of Public Health's new cutting-edge training program.

The inaugural Public Health and AI Summer School kicked off June 9 and included four days of workshops and hands-on learning to explore how artificial intelligence can be used in the public health arena with readily available tools such as those that allow researchers to collect digital data from multiple sources and analyze it quickly.

Image

Along with the Zuckerman College of Public Health, the collaborative effort included the Global Health Institute and the Data Science Institute.

U of A experts including Onicio Leal Neto, Dr. Marvin Slepian, and Nirav Merchant introduced AI basics and covered a wide range of topics, from ethical and legal considerations to the use of machine learning in patient care and its challenges in government use. The goal was for participants – whether they had previous AI experience or none at all – to come away with an understanding of AI's relevance in and its applications for public health practice, including surveillance, forecasting, risk prediction and administrative tasks. 

Speakers also included industry leaders, as well as county and state health officials. 

A highlight of the program was the ability for participants to break into groups at the end of each day's session to put AI tools to use.

"We're training the next generation of public health professionals to use AI ethically and effectively to benefit all communities," said Dr. Iman Hakim, dean of the Zuckerman College of Public Health. "Our college is already using AI in research and programs, giving students hands-on experience. We're providing the tools, the experience and the space to explore how AI can help us do what we do best – protect and improve lives."

Training was in line with new initiative

The Zuckerman College of Public Health recently launched an AI for Public Health Initiative, which, in addition to the summer program, includes new courses and research projects.

The effort aligns with the U of A's strategic imperatives, which include a $20 million investment in areas of strategic importance not just for the state but also the nation, such as enhancing biomedical discoveries and health care outcomes with artificial intelligence and machine learning. Additionally, the U of A recently brought in its first chief AI and data science officer, David Ebert. An internationally recognized expert in artificial intelligence and data science, Ebert will craft the university's AI and data science vision.

The summer program is the brainchild of Leal Neto, an assistant research professor of digital epidemiology at the Zuckerman College of Public Health. Leal Neto said he was excited by the amount of interest in the event, which attracted faculty, staff and students from different disciplines, health professionals and even people from out of state. In fact, more than 3,100 people from across the globe visited the program's registration website. Initially, the workshop was going to be capped at 25 participants, but it ended up with 52 attendees and a waiting list.

"We are working hard to bring the necessary training for students and also for people from outside to understand the importance of AI and machine learning for the public health function, but with the perspective of public health practitioners and professionals," Leal Neto said.

AI is evolving and changing rapidly, and students need to be taught the most relevant, up-to-date practices, he said.

"We are doing our best to get our students in a better position to be in the job market as soon as they get out of our programs," he said. 

Building blocks for future use

The theme throughout the week was that AI can be incredibly useful, but it requires transparency and must be used responsibly, with informed consent and human oversight, in order for it to be used safely, effectively and equitably.

Image

"We are entering an environment which is highly nonpoliced," Slepian said. "The onus is on us to put the regulatory upfront. We need the law upfront, so we don't have hassles later." 

"Our goal is to empower (the participants) and introduce them to pragmatic tools and resources," said Merchant, director of the Data Science Institute and a member of the BIO5 Institute. "We told them, 'You're going to take home your powerful AI toolbox, and you have the freedom to customize and apply it to your processes. It's like working with Lego blocks – you can build exactly what you need.' We're sharing with them the best practices grounded in public health."

Slepian, a Regents Professor of medicine, medical imaging and surgery at the College of Medicine – Tucson and of biomedical engineering at the College of Engineering, said AI is the buzzword for 2025. He noted that pioneers in artificial intelligence were awarded two Nobel Prizes last year, and AI is already widely used in radiology where it's been shown to improve detection of breast cancer. 

"It's not going away. It's here to stay," said Slepian, who runs the Arizona Center for Accelerated Biomedical Innovation and is also a member of BIO5. "The key to AI is knowing what's under the hood."

Each day of the seminar ended with time set aside for participants to put into practice what they'd learned. One project called for using free AI tool Jotform to create an AI public health agent, build its knowledge base and evaluate its responsiveness as well as its clarity and relevance. Once built, other participants tested the agents by asking questions. Groups trained their chatbots on specific health subjects ranging from pregnancy, human papillomavirus vaccines and rabies exposure. 

Christine Girard, a senior lecturer at the Zuckerman College of Public Health who works in Phoenix, said she attended the summer program because students are already using AI tools. As someone without a background in AI, Girard wanted to learn in an academic setting so she can guide students to using the best tools and to do so responsibly. 

"This week has been very special, and I think the number of us who have come together across different departments, and even sectors of health, and health sciences and data science has made our experience more robust than it would have been if it was just one group of people," she said.

Girard said as a brand-new user, she found all the information that was shared very useful. She hoped the program would be the start of a continued collaboration of sharing and exploring AI insights with each other.

"It would be great to communicate with one another to develop not only a community of learning but also a community of practice to continue to work together and expand the knowledge and experience that we've had this week," she said.

Leal Neto said he expects to create not only additional AI summer programs but also hopes to hold an international summit on the topic.

Katy Smith, a science writer in the U of A Health Sciences Office of Communications, contributed reporting. A version of this article originally appeared on the U of A Health Sciences website.