1Timothy Sullivan [99]United States (Single Center)2018Hospital settingEHR data, Klebsiella pneumoniae bacteremia casesMultiple logistic regressionAUROC: 0.731, Sensitivity: 73%, Specificity: 59%, PPV: 16%, NPV: 95%Klebsiella pneumoniae (Carbapenem-resistant)2Ariane Khaledi [71]Germany, Spain2020Clinical settings, multicenterWhole genome sequencing (WGS), transcriptomic data, gene presence/absence, expression profilesMachine Learning (unspecified classifiers)Sensitivity: 0.8–0.9, Predictive values: >0.9Pseudomonas aeruginosa (Carbapenem-resistant)3Ed Moran [97]United Kingdom (Single Center)2020Hospital settingBlood and urine cultures, demographics, microbiology and prescribing dataXGBoostAUROC: 0.70, Point-scoring tools: AUROC 0.61 to 0.67, estimated reduction in broad-spectrum antibiotic use by 40%Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa4Ryan J. McGuire [80]United States (Single Center)2021Tertiary-care academic medical centerDemographics, medications, vital signs, procedures, lab results, culturesExtreme gradient boosting (XGBoost)AUROC: 0.846, Sensitivity: 30%, PPV: 30%, NPV: 99%Carbapenem-resistant bacteria5Maddalena Giannella [90]Multinational2021Liver transplantation units (multicenter)Demographics, clinical data, mechanical ventilation, acute renal injury, surgical reinterventionMultivariable logistic regression, Fine-Gray subdistribution hazard modelAUROC: 74.6 (derivation), AUROC: 73.9 (bootstrapped validation), Brier Index: 16.6Carbapenem-resistant Enterobacteriaceae (CRE)6Qiqiang Liang [79]China (Single Center)2022Intensive care unit (ICU)Demographics, screening records, clinical data, vitalsRandom forest, XGBoost, decision tree, logistic regressionAUROC: 0.91 (random forest), 0.89 (XGBoost, decision tree), 0.78 (logistic regression)Carbapenem-resistant Gram-negative bacteria (CRGNB)7Maristela Pinheiro Freire [91]Brazil, Italy2022Liver transplantation units (multicenter)Antibiotic use, hepato-renal syndrome, CLIF-SOFA scores, cirrhosis complicationsMachine learning (unspecified)Sensitivity: 66%, Specificity: 83%, NPV: 97%Carbapenem-resistant Enterobacterales (CRE)8Çaǧlar Çaǧlayan [92]United States (Single Center)2022Intensive care unit (ICU)EHR, MDRO screening program, sociodemographic and clinical factorsLogistic regression (LR), random forest (RF), XGBoostSensitivity: VRE 80%, CRE 73%, MRSA 76%, MDRO 82%; Specificity: VRE 66%, CRE 77%, MRSA 59%, MDRO 83%MRSA, VRE, Carbapenem-resistant Enterobacteriaceae (CRE)9Qiqiang Liang [63]China (Single Center)2024Intensive care unit (ICU)Demographics, mechanical ventilation, invasive catheterization, carbapenem use historyRandom forest, XGBoost, SVMAUROC: random forest 0.86, XGBoost (infection): 0.86, SVM: 0.88, RF (CRGNB): 0.87Carbapenem-resistant Gram-negative bacteria (CRGNB)10Yun Li [65]China/USA2024Intensive care unit (ICU)Electronic health record data (PLAGH-ICU, MIMIC-IV)Machine learning modelsAUROC: 0.786 (PLAGH-ICU), 0.744 (MIMIC-IV)Multidrug-resistant organisms (MDRO), including carbapenem-resistant species11Bing Liu [64]China (Single Center)2024Multiple hospital settingsWhole-genome sequencing (WGS) data, metagenomic sequencing (MGS), genomic featuresMachine learning (unspecified algorithms)AUROC: 0.906 (IPM), 0.925 (MEM), PPV: 0.897 (IPM), 0.889 (MEM)Pseudomonas aeruginosa (Carbapenem-resistant)