Large Language Model-Based Clinical Decision Support for Antibiotic Selection and Dose Recommendation in…

AI / informaticsJMIR medical informatics· Aug 2026

Large Language Model-Based Clinical Decision Support for Antibiotic Selection and Dose Recommendation in Hospitalized Patients With Pneumonia: Multicenter Retrospective Study.

Yang Zhang, Li Li, Chunting Tan … Honglei Liu

Clinical Decision SupportAI / informaticsQ1IF 5.8

Constrained large language models pipeline matched antibiotic selection with high accuracy in external validation.

Large Language Model-Based Clinical Decision Support for Antibiotic Selection and Dose Recommendation in Hospitalized Patients With Pneumonia: Multicenter Retrospective Study.

Yang Zhang … Honglei Liu

JMIR medical informatics · Aug 2026

Background

Pneumonia is a common infectious disease, and antibiotic treatment in hospitalized patients must balance efficacy, safety, and resistance risk.

Purpose

This study aimed to develop and externally validate a constrained large language models-based clinical decision support pipeline for antibiotic selection and dose recommendation in hospitalized patients with pneumonia.

Methods

The development cohort included 233 patients, and the external validation cohort included 98 patients.

n = 233 patients

Results

external validation f1 score, near 1.0 means near-perfect antibiotic matching to clinician choices

0.811F1-sco

Internal test set

0.8605F1-sco

External validation se

More results

The system also provided traceable evidence and rule trigger information to support clinician review.

“

Conclusion

A constrained, retrieval-augmented large language models pipeline improved the consistency and interpretability of antibiotic selection and dose recommendation for hospitalized patients with pneumonia and provided preliminary evidence of cross-site generalizability.

Large Language Model-Based Clinical Decision Support for Antibiotic Selection and Dose Recommendation in Hospitalized Patients With Pneumonia: Multicenter Retrospective Study.

Large Language Model-Based Clinical Decision Support for Antibiotic Selection and Dose Recommendation in Hospitalized Patients With Pneumonia: Multicenter Retrospective Study.

Related papers