Israeli study finds custom prescriptions can slow antibiotic resistance

AI could be used to decrease the risk of antibiotic resistance by 50 percent, according to new research by the Technion and Maccabi KSM Research and Innovation Center.

An illustrative picture of antibiotic resistance tests (photo credit: Wikimedia Commons)
An illustrative picture of antibiotic resistance tests
(photo credit: Wikimedia Commons)

Perhaps the single most important development of the modern era, antibiotics save untold billions of human lives around the world. Unfortunately, the more they are used, the less effective they become.

AI, however, could be used to decrease the risk of antibiotic resistance by 50 percent, researchers from the Technion and Maccabi KSM Research and Innovation Center found in a peer-reviewed study published in the journal Science on Thursday. The study was supported by the US National Institutes of Health (NIH), the Israel Science Foundation in the Israel Precision Medicine Partnership program, the European Research Council (ERC), the Ernest and Bonnie Beutler Research Program of Excellence in Genomic Medicine, the D. Dan & Betty Kahn Foundation and the Wellcome Trust.

Over time, as bacteria multiply and evolve, they develop mutations, including resistance to antibiotics. This effectively means diseases that are normally treatable become untreatable. However, the Israeli researchers discovered a way to personalize antibiotic prescriptions based on each patient's medical history.

They studied antibiotic resistance in patients with wound and urinary tract infections and found that, in most cases, the resistance was developed as a result of reinfection by antibiotic-resistant bacteria already present in the patient's body. Thus, the researchers developed a machine-learning algorithm to predict the risk of reinfection with resistant bacteria based on their past history of infection.

 Nitzan Zohar/Technion spokesperson's office (credit: NITZAN ZOHAR/TECHNION SPOKESPERSON'S OFFICE) Nitzan Zohar/Technion spokesperson's office (credit: NITZAN ZOHAR/TECHNION SPOKESPERSON'S OFFICE)

Dr. Mathew Stracy, the primary author of the study, said: "We found that the antibiotic susceptibility of the patient’s past infections could be used to predict their risk of returning with a resistant infection following antibiotic treatment. Using this data, together with the patient’s demographics like age and gender, allowed us to develop the algorithm."