AI could be key to quicker disease detection and treatment, expert says

Five years after the COVID-19 pandemic hit the nation, many people are starting to believe the risk of outbreaks could be a thing of the past. But one expert says public health agencies know better than to let their guard down. 

“What we’ve seen historically is that there’s a large-scale outbreak about every seven to 10 years,” said Brian Dixon, Director of the Regenstrief Institute's Clem McDonald Center for Biomedical Informatics. He noted that public health officials are thinking, “What is going to be the next outbreak that turns into a pandemic? Where is that novel strain of disease going to come from, and how will it present, and where will it start?”

So far this year, for instance, 10 states have already reported measles outbreaks as the number of cases nationwide is nearly 900. In recent months, public health officials have long raised concerns over the infections of a new bird flu outbreak that is spreading to previously uninfected species like dairy cows. Some states and towns have also flagged the growing threat of mosquito-borne illnesses, like West Nile virus, as climates warm. 

Other health risks like cardiovascular death rates have also increased — up 4% in 2024 compared to the previous year — as heart disease remains a leading cause of fatalities in the nation, according to the American Heart Association. The rate of cancer diagnoses, particularly among women and people under the age of 50, have also been on the rise in recent years.  

To address such threats, public health departments have long made use of data systems and infrastructure that leverage electronic health records, wastewater surveillance insights or laboratory test results from the health care sector, Dixon said.

But as many agencies grapple with legacy tech or even still rely on manual- and paper-based data collection, officials could struggle to adequately track and manage future outbreaks, putting lives and already strapped government resources at risk. 

A new partnership between the University of Pittsburgh and Leidos aims to advance public health disease monitoring by developing artificial intelligence-enabled detection tools. 

Leidos has invested $10 million to work with the university’s Computational Pathology and AI Center of Excellence to explore how AI can help more efficiently identify diseases, reduce diagnosis times and inform care and treatment management, the company announced last month. 

The partnership ultimately plans to deploy such solutions to the public and private health sector for widespread improvement of health care delivery. Under the initiative, for instance, the organizations will research and develop AI-enabled digital pathology solutions, such as advanced technology for scanning tissue samples and analyzing medical images.

AI’s use in public health is still in its early stages, Dixon said. But agencies should consider turning to more innovative solutions like artificial intelligence and automation, he said, which could be key to identifying, tracking and treating disease and other health concerns in communities. 

The Oklahoma Department of Health, for instance, announced it is replacing its legacy public health surveillance system with an AI-enabled platform earlier this year. The new system aims to streamline and expedite the reporting, tracking and managing of infectious diseases across the state. 

The need for quicker and more effective data monitoring with innovative approaches like AI and other advanced computing is particularly crucial amid declining vaccination rates, particularly among school-aged children, as misinformation and vaccine skepticism contribute to Americans’ hesitancy toward immunization. 

According to Dixon, enough adults are forgoing vaccines for themselves and their children that “we’ve seen [immunization] levels fall below what we know is necessary for herd immunity.” 

“Monitoring health conditions at the population level remains critical,” he said. “Humans can remain in contact with one another, they're in contact with animals and they're in contact with the environments, and because of that, they are exposing themselves constantly to different pathogens.”

AI can help agencies collect and process more health data to identify, and even predict, where and when potential outbreaks are emerging, Dixon said. Those insights can inform health officials where to concentrate funding for resources — such as testing devices, vaccinations and other treatment services — to further contain the impact of diseases in communities. 

Earlier detection and treatment, he explained, hold potential to save governments money by preventing certain cases from reaching emergency levels, which are often more expensive for residents and governments to cover. Plus, more effective resource allocation can help reduce the extra spending of public dollars.

“Public health’s job is to keep the whole population safe, and, in order to do that, you really want to have a handle on what disease is in your community and … if it’s spreading,” Dixon said.