09/11/2026
Smarter, more personalized antibiotic selection.
University of Iowa researchers developed a new machine learning model to predict antibiotic resistance. Led by Michihiko Goto, MD, MS, this work could someday help clinicians choose the right treatment faster.
When a patient develops a serious bacterial infection, doctors often need to begin treatment before lab results show which antibiotics will work. Factors like a patient’s previous infections, recent antibiotic use, or other health conditions can all affect treatment effectiveness.
The researchers' model looks for patterns in a patient’s medical history that could indicate antibiotic resistance, which could make those early decisions more accurate. Future studies will expand to other organisms and explore how the model could fit into everyday clinical practice.
“The information is already in the medical record,” Goto says. “The question is how to put it to work for every prescriber, every time.”