biostudies-literatureUnknown38Cai C<h4>Introduction</h4>Carbapenem-resistant <i>Klebsiella pneumoniae</i> (CRKP) poses a significant public health threat. Rapid detection of CRKP and its resistance mechanisms is essential for optimizing antibiotic therapy and infection control. However, clinical implementation faces several challenges.<h4>Methods</h4>Machine learning classifiers were applied using MALDI-TOF MS data to discriminate KPC-type, NDM-type CRKP, and carbapenem-susceptible strains (CSKP). Model performance was validated across platforms and strain collections. SHapley Additive exPlanations (SHAP) analysis and phylogenetic reconstruction were used to interpret feature contributions and genetic determinants.<h4>Results</h4>Significant spectral divergence was observed among <i>K. pneumoniae</i> phenotypes, particularlJournal of mass spectrometry and advances in the clinical lab71-80https://www.ebi.ac.uk/biostudies/studies/S-EPMC12648985biostudies-literatureInterpretable machine learning of clinical MALDI-TOF spectra discriminates carbapenem-resistant &lt;i&gt;Klebsiella pneumoniae&lt;/i&gt; while revealing phylogenetic heterogeneity that limits model generalizability.PMC12648985Cai CFang ZYuan PZeng NZou MChen MZhou HZhong LChen DfalseInterpretable machine learning of clinical MALDI-TOF spectra discriminates carbapenem-resistant &lt;i&gt;Klebsiella pneumoniae&lt;/i&gt; while revealing phylogenetic heterogeneity that limits model generalizability.<h4>Introduction</h4>Carbapenem-resistant <i>Klebsiella pneumoniae</i> (CRKP) poses a significant public health threat. Rapid detection of CRKP and its resistance mechanisms is essential for optimizing antibiotic therapy and infection control. However, clinical implementation faces several challenges.<h4>Methods</h4>Machine learning classifiers were applied using MALDI-TOF MS data to discriminate KPC-type, NDM-type CRKP, and carbapenem-susceptible strains (CSKP). Model performance was validated across platforms and strain collections. SHapley Additive exPlanations (SHAP) analysis and phylogenetic reconstruction were used to interpret feature contributions and genetic determinants.<h4>Results</h4>Significant spectral divergence was observed among <i>K. pneumoniae</i> phenotypes, particularl2025-01-01T00:00:00Z2025 Dec2026-05-19T03:27:40.784Z2026-05-19T03:12:24.755ZS-EPMC126489854131146210.1016/j.jmsacl.2025.11.003