EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access · CC BY 4.0

Personalized Antibiogram: A Novel Multitask Machine Learning Framework for Simultaneous Prediction of Antimicrobial Resistance Profile With Enhanced Detection of Carbapenem Resistance in Enterobacteriaceae

Michihiko Goto iD, Anindita Bandyopadhyay iD, Qianyi Shi, Yaohua Wang iD, Eli N Perencevich, David Hernandez, W Nick Street

Clinical Infectious Diseases · 2026

DOI: 10.1093/cid/ciag027 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Clinical Infectious Diseases, 2026. Verfasst von Michihiko Goto und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, Iowa, USA. Hauptthemen laut MEDLINE: Anti-Bacterial Agents, Machine Learning, Carbapenems, Carbapenem-Resistant Enterobacteriaceae. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: AHRQ HHS, Agency for Healthcare Research and Quality.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Conventional hospital antibiograms summarize aggregated resistance rates, limiting their utility for individualized antimicrobial selection. Existing statistical and machine learning models predict each phenotype separately, ignoring correlations among resistance profiles. We developed novel multitask extreme gradient boosting (XGBoost) models utilizing structured data in electronic health records (EHRs) to predict resistance to 8 antimicrobial classes simultaneously and evaluated their performance within the Veterans Health Administration (VHA).

Methods

We conducted a retrospective multicenter study of Escherichia coli and Klebsiella spp. isolates collected at 127 hospitals and >1400 clinics from January 2017 to September 2024. Data from January 2017 to September 2023 were used for model development, while data from October 2023 to September 2024 were used for simulated prospective testing. Model performances were compared to hospital antibiograms and single-target XGBoost models.

Results

The training cohort included 536 252 E. coli and 246 898 Klebsiella spp. isolates; the test cohort included 75 138 and 38 015 isolates, respectively. On the test data, the multitask model achieved overall areas under the receiver operating characteristic curve (AUROCs) of 0.779 (E. coli) and 0.810 (Klebsiella spp.), with good to excellent per-class performance (AUROC range, 0.743–0.847). A multitask approach improved calibration and decreased false-negative rates for carbapenem resistance while predicting individualized resistance probabilities for all target antimicrobials simultaneously (“personalized antibiograms”).

Conclusions

A multitask XGBoost framework can accurately predict individualized resistance profiles for common Gram-negative pathogens, outperforming conventional antibiograms and single-target models. Personalized antibiograms may enhance the selection of empiric therapy, including the detection of carbapenem resistance in low-endemicity settings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Anti-Bacterial AgentsMachine LearningCarbapenemsCarbapenem-Resistant EnterobacteriaceaeDrug Resistance, BacterialEnterobacteriaceaeHumansMicrobial Sensitivity TestsBoosting Machine Learning AlgorithmsRetrospective StudiesEscherichia coliKlebsiellaPredictive Learning ModelsEnterobacteriaceae Infections

Stichwörter der Autor:innen

Gram-negative rodsantimicrobial resistance predictionelectronic health record datamachine learning

Substanzen

Anti-Bacterial AgentsCarbapenems

Autor:innen und Institutionen

  • Department of Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, Iowa, USA.
  • Center for Access & Delivery Research and Evaluation (CADRE), Iowa City Veterans Affairs Health Care System, Iowa City, Iowa, USA.
  • Department of Business Analytics, University of Iowa Tippie College of Business, Iowa City, Iowa, USA.

ORCID-Profile: Michihiko Goto, Anindita Bandyopadhyay, Yaohua Wang

Förderung

  • AHRQ HHS
  • Agency for Healthcare Research and Quality

Zitieren

APA

Goto, M., Bandyopadhyay, A., Shi, Q., Wang, Y., Perencevich, E. N., Hernandez, D., & Street, W. N. (2026). Personalized Antibiogram: A Novel Multitask Machine Learning Framework for Simultaneous Prediction of Antimicrobial Resistance Profile With Enhanced Detection of Carbapenem Resistance in Enterobacteriaceae. Clinical Infectious Diseases. https://doi.org/10.1093/cid/ciag027

Vancouver

Goto M, Bandyopadhyay A, Shi Q, Wang Y, Perencevich EN, Hernandez D, et al. Personalized Antibiogram: A Novel Multitask Machine Learning Framework for Simultaneous Prediction of Antimicrobial Resistance Profile With Enhanced Detection of Carbapenem Resistance in Enterobacteriaceae. Clinical Infectious Diseases. 2026. doi:10.1093/cid/ciag027

BibTeX
@article{Goto2026,
  author = {Goto, Michihiko and Bandyopadhyay, Anindita and Shi, Qianyi and Wang, Yaohua and Perencevich, Eli N and Hernandez, David and Street, W Nick},
  title = {Personalized Antibiogram: A Novel Multitask Machine Learning Framework for Simultaneous Prediction of Antimicrobial Resistance Profile With Enhanced Detection of Carbapenem Resistance in Enterobacteriaceae},
  journal = {Clinical Infectious Diseases},
  year = {2026},
  doi = {10.1093/cid/ciag027}
}
Kostenlos starten →