Use of Artificial Intelligence in Current Fight Against Antimicrobial Resistance

Area of Science:

• Pharmacology
• Infectious Diseases
• Computational Biology

Background:

• Antimicrobial resistance (AMR) is a growing global health crisis, projected to exceed cancer mortality by 2050.
• Optimizing antimicrobial dosing is crucial for effective treatment and resistance mitigation.
• Traditional pharmacokinetic (PK) modeling methods face limitations in complexity and mechanistic requirements.

Purpose of the Study:

• To review the application of artificial intelligence (AI) and machine learning (ML) in optimizing antimicrobial dosing.
• To highlight AI/ML's potential in managing pharmacokinetic and pharmacodynamic variability.
• To discuss the advantages and challenges of integrating AI/ML into clinical practice for combating AMR.

Main Methods:

• Leveraging large datasets for accurate drug exposure prediction using ML models.
• Refining sampling strategies and enabling real-time dose adjustments via therapeutic drug monitoring.
• Comparing AI/ML model performance against traditional methods in case studies.

Main Results:

• AI and ML models demonstrate potential to match or exceed traditional methods in achieving therapeutic targets.
• Case studies with ganciclovir, vancomycin, and daptomycin show AI/ML efficacy in managing variability and toxicity.
• AI/ML offers dynamic adaptability and precision in antimicrobial dosing strategies.

Conclusions:

• AI and ML present a transformative approach to precision medicine in infectious disease management.
• Addressing challenges like data quality and interpretability is key for widespread AI adoption.
• Future research should focus on multi-omics integration and developing clinical decision-support tools to combat AMR effectively.