by The team, led by Dr. Jacob Scott, published their findings in the Proceedings of the National Academy of Sciences.
Antibiotics have significantly contributed to increasing the average lifespan in the US by nearly ten years. However, their effectiveness is waning due to overuse and the emergence of antibiotic-resistant bacteria. Health agencies worldwide warn that if we don’t change our approach to treating bacterial infections, more people will die from antibiotic-resistant infections than from cancer by 2050.
Bacteria reproduce rapidly, allowing them to produce mutant offspring. Overusing antibiotics provides an opportunity for bacteria to develop resistance, leaving only the stronger strains that are resistant to treatment.
To address this issue, physicians are employing antibiotic cycling, which involves rotating between different antibiotics over specific time periods. This strategy limits the amount of time bacteria have to develop resistance to any one class of antibiotic and can even make bacteria more susceptible to other antibiotics.
However, there is a lack of standardization in antibiotic cycling between hospitals, making it challenging to determine the best antibiotic to use, for how long, and in what order.
To tackle this problem, study co-author Dr. Jeff Maltas, a postdoctoral fellow at Cleveland Clinic, and first author Davis Weaver, Ph.D., a medical student, used data-driven models to predict drug cycling regimens that minimize antibiotic resistance and maximize antibiotic susceptibility.
Dr. Weaver led the application of reinforcement learning to the drug cycling model, which enables a computer to learn from its mistakes and successes to determine the best strategy to complete a task. This study marks one of the first instances of reinforcement learning being applied to antibiotic cycling regimens.
Reinforcement learning is an ideal approach because determining the bacteria’s growth rates is relatively straightforward. There is also room for human variations and errors, making it unnecessary to measure growth rates precisely every time.
The research team’s AI was able to identify the most efficient antibiotic cycling plans to treat multiple strains of E. coli and prevent drug resistance. Dr. Maltas emphasizes that AI can support complex decision-making, such as calculating antibiotic treatment schedules.
Furthermore, the team’s AI model can inform how hospitals approach antibiotic treatment beyond individual patients. The researchers are also working to expand their work beyond bacterial infections to tackle other deadly diseases that can develop treatment resistance.
“This idea isn’t limited to bacteria,” Dr. Weaver explains. “In the future, we believe these types of AI can be used to manage drug-resistant cancers as well.”
Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it
