Multi-label classification to predict antibiotic resistance from raw clinical MALDI-TOF mass spectrometry data

This study explores multi-label classification as a novel approach to predict antibiotic resistance across four bacteria: E. coli, S. aureus, K. pneumoniae, and P. aeruginosa. Using datasets from the DRIAMS repository, the multi-label approach outperforms traditional single-label models. The multi-label framework accurately captures real-world scenarios, and the models’ generalizability and robustness were confirmed. This suggests multi-label classification can enhance predictive accuracy in AMR research.