MALDI-TOF MS platform combined with machine learning to establish a model for rapid identification of methicillin-resistant Staphylococcus aureus

MALDI-TOF MS is an effective potential tool to distinguish between MSSA and MRSA . By combining the ClinProTools3.0 software and manual grouping intervention, we proposed a model optimization method for the first time. The cross validation of the model increased from 95.82% to 96.68%, and the accuracy of the model increased from 88.89% to 91.98%. Finally, we reported nine characteristic peaks of rapid detection of MRSA. • MALDI-TOF MS combined with ClinProTools is a potential tool for rapid detection of MRSA isolates. • The detection model of MRSA can be optimized by establishing multiple databases with different groups. • The model based on genetic algorithm (GA) shows great performance for MRSA detection.