Artificial Intelligence able to identify potentially life-saving antibiotics

Researchers from University Hospital Basel (USB) and the Department of Biosystems Science and Engineering (D-BSSE) at the Swiss Federal Institute of Technology in Zurich (ETH) in Basel are using innovative computer algorithms in the fight against antibiotic-resistant bacteria. With this new method, the researchers can identify characteristics of antibiotic resistance in germs up to 24 hours earlier. This means that it is clear sooner rather than later which antibiotics can still help patients. The cantons of Basel-Stadt and Basel-Landschaft offered financial support for this research project.

“In cases of severe infection, the time until optimal therapy can be the difference between life and death. A quick and precise diagnosis is extremely important here”, comments Professor Adrian Egli, Head of the Clinical Bacteriology department at USB and study director, in a press release.

Larger clinical study will follow

The study team linked 300,000 mass spectrometry data samples from individual bacteria from four laboratories in north-western Switzerland with the results of antibiotic resistance tests in the laboratory. “Intelligent computer algorithms look for patterns in the data that distinguish bacteria with and without resistance from one another”, explains Caroline Weis, PhD student at the D-BSSE at ETH Zurich in Basel and lead author of the study.

The researchers have published their method in the latest edition of the specialist journal “Nature Medicine”. It must now be confirmed in a larger clinical study carried out as part of everyday hospital operations. Plans are already in place for this to take place at USB.