Cleveland Clinic study finds artificial intelligence can develop treatments to prevent ‘superbugs’

CLEVELAND, Ohio — Researchers at the Cleveland Clinic are using artificial intelligence to create the next defense against superbugs — bacteria that have developed a resistance to antibiotics — by using computers to model the use of antibiotics and recommend better choices.

Bacteria replicate quickly, often evolving mutations that help them adapt and survive. Overuse of a single antibiotic gives subsequent generations of bacteria the opportunity to develop the ability to resist treatment. Over time, the antibiotics kill all the susceptible bacteria, leaving behind only the resistant progeny.

To combat this phenomenon, a common strategy is rotating between different antibiotics over specific time periods, a process called antibiotic cycling.

By alternating between different drugs, bacteria have less opportunity to evolve resistance to any one class of antibiotic and can even increase their vulnerability to the drugs.

Cleveland researchers Davis Weaver and Jeff Maltas are attempting to optimize the antibiotic cycling by teaching a computer to learn from its mistakes and successes. Their findings were published in April in the journal PNAS, and are among the first to apply reinforcement learning to antibiotic cycling Weaver and Maltas said.

“Drug cycling shows a lot of promise in effectively treating diseases,” said Weaver a medical student at Case Western Reserve School of Medicine, and lead author of the study. “The problem is that we don’t know the best way to do it. Nothing’s standardized between hospitals for which antibiotic to give, for how long and in what order.”

“Reinforcement learning is an ideal approach because you just need to know how quickly the bacteria are growing, which is relatively easy to determine,” explained Weaver. “There’s also room for human variations and errors. You don’t need to measure the growth rates perfectly down to the exact millisecond every time.”

Weaver worked together with the study’s co-author Jeff Maltas, a postdoctoral fellow at Cleveland Clinic. Maltas uses computer models to predict how resistance to one antibiotic can make it weaker to another. They wanted to see if the models could predict drug cycling regimens that both minimized antibiotic resistance and maximized and antibiotic susceptibility.

The research team’s AI was able to figure out the most efficient antibiotic cycling plans to treat multiple strains of E. coli and prevent drug resistance. The study shows that AI can support complex decision-making like calculating antibiotic treatment schedules, Maltas said.

It’s a real-world solution to a problem of growing importance said Dr. Jacob Scott, who oversaw the work in his laboratory at the Cleveland Clinic.

“Health agencies worldwide agree that we’re entering a post-antibiotic era,” explains Dr. Scott. “If we don’t change how we go after bacteria, more people will die from antibiotic-resistant infections than from cancer by 2050.”

In addition to managing an individual patient’s infection, the team’s AI model can help hospitals determine the best way treat other types of infections as well said Weaver.

“This idea isn’t limited to bacteria, it can be applied to anything that can evolve treatment resistance,” he says. “In the future we believe these types of AI can be used to manage drug-resistant cancers, too.”