EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

abx_amr_simulator: A simulation environment for antibiotic prescribing policy optimization under antimicrobial resistance

Joyce Lee, Seth Blumberg

SoftwareX · 2026-09 · 35 · 102872

DOI: 10.1016/j.softx.2026.102872 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SoftwareX, 2026-09. Verfasst von Joyce Lee und 1 weiteren Autor:innen. Die Arbeit nennt 9 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Centers for Disease Control and Prevention.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Antimicrobial resistance (AMR) is a critical global health threat, yet quantitatively evaluating the impact of antibiotic stewardship interventions remains a challenge due to incomplete or biased real-world data. While pathogen-level simulations exist, there are few tools focused on patient-level antibiotic prescribing decisions. We introduce abx_amr_simulator , a customizable Python framework designed to allow users to configure a simplified and idealized simulated environment that still captures the core dynamics of real-world antibiotic prescribing and antimicrobial resistance responses. This simulator therefore frames antibiotic prescribing as solving a sequential decision-making task under uncertainty, where the simulator allows the user to specify the nature and degree of uncertainty. The package provides a modular environment that allows users to configure patient heterogeneity, communal resistance dynamics, and information degradation (e.g., noise, bias, and delays). By conforming to the Gymnasium API, the simulator allows researchers to quantitatively evaluate and compare a wide range of prescribing strategies, from simple heuristic rules to reinforcement learning agents. To support complex long-horizon tasks, the package also includes reference implementations for hierarchical reinforcement learning. abx_amr_simulator offers a versatile, “turnkey” interface—accessible via GUI, CLI, or custom Python subclasses—enabling clinicians and computational researchers alike to benchmark stewardship policies in a controlled, rigorous environment.

Förderung

  • Centers for Disease Control and Prevention

Zitieren

APA

Lee, J., & Blumberg, S. (2026). abx_amr_simulator: A simulation environment for antibiotic prescribing policy optimization under antimicrobial resistance. SoftwareX, 35, 102872. https://doi.org/10.1016/j.softx.2026.102872

Vancouver

Lee J, Blumberg S. abx_amr_simulator: A simulation environment for antibiotic prescribing policy optimization under antimicrobial resistance. SoftwareX. 2026;35:102872. doi:10.1016/j.softx.2026.102872

BibTeX
@article{Lee2026,
  author = {Lee, Joyce and Blumberg, Seth},
  title = {abx_amr_simulator: A simulation environment for antibiotic prescribing policy optimization under antimicrobial resistance},
  journal = {SoftwareX},
  year = {2026},
  volume = {35},
  pages = {102872},
  doi = {10.1016/j.softx.2026.102872}
}
Kostenlos starten →