July 15, 2025Open Access

Early detection of ampicillin susceptibility in Enterococcus faecium with MALDI-TOF MS and machine learning

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Machine learning techniques identify ampicillin susceptibility in Enterococcus faecium, suggesting optimized treatment strategies.

LightGBM outperformed logistic regression in detecting ampicillin-susceptible isolates in both TUM and MS-UMG datasets.
The analysis showed AUPRC values of 0.907 for LightGBM vs 0.902 for logistic regression in the TUM dataset.
External validation indicated good transferability of the models, with AUPRCs of 0.784 and 0.804 across datasets.
Identification of specific spectral features in MALDI-TOF data supports the potential for rapid antibiotic susceptibility testing.

Pichl et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark