Artificial intelligence-driven approaches in antibiotic stewardship programs and optimizing prescription practices: A...

Area of Science:

• Infectious Diseases
• Medical Informatics
• Computational Biology

Background:

• Antimicrobial resistance (AMR) is a critical global health challenge requiring optimized antibiotic use.
• Antimicrobial stewardship programs (ASPs) are vital for combating AMR.
• Artificial intelligence (AI) and machine learning (ML) offer novel approaches to enhance ASPs.

Purpose of the Study:

• To systematically review the application of AI-driven tools in ASPs.
• To evaluate the methodologies, outcomes, and challenges of AI in antimicrobial stewardship.
• To assess the impact of AI on antibiotic prescription accuracy, resistance prediction, and dosage optimization.

Main Methods:

• Systematic review of studies from PubMed, Scopus, Web of Science, and Embase.
• Keywords: "AI" and "antibiotic."
• Inclusion criteria: studies using AI/ML in ASPs for empirical selection, dose adjustment, and adherence; risk of bias assessed using Newcastle Ottawa Scale.

Main Results:

• AI/ML models demonstrate effectiveness in optimizing empirical antibiotic selection and predicting resistance.
• Studies confirm AI's role in enhancing therapy appropriateness, potentially reducing mortality.
• Machine learning shows promise in optimizing antibiotic dosing, notably for vancomycin.

Conclusions:

• AI and ML significantly enhance antibiotic stewardship through optimized interventions, selection, prediction, and dosing.
• AI-driven ASPs have the potential to improve healthcare outcomes and combat AMR.
• Disparities exist in AI implementation between high-income and low/middle-income countries, highlighting structural challenges.