Retrospective validation study of a machine learning-based software for empirical and organism-targeted antibiotic therapy selection

A study in a 12-hospital Spanish institution assessed the efficacy of iAST, a machine-learning-based software that offers empirical and organism-targeted antibiotic recommendations. The study found that all three iAST recommendations were non-inferior to doctor prescriptions. The overall success rate for empirical treatment was 68.93%, while the success rate for organism-targeted therapy was 84.16%. iAST showed a greater propensity to recommend access antibiotics, fewer watch antibiotics, and higher reserve antibiotics. This study highlights iAST’s potential to promote effective antibiotic stewardship.