Machine-learning-enabled modeling of pharmacokinetics and pharmacodynamics

Area of Science:

• Pharmacometrics
• Computational Biology
• Machine Learning

Background:

• Computational modeling is crucial for pharmacokinetic (PK) and pharmacodynamic (PD) assessments.
• The pharmacometrics field is integrating machine learning (ML) and artificial intelligence (AI) for enhanced pharmacology.
• AI/ML offers potential benefits like faster assessments, reduced costs, and improved patient safety.

Purpose of the Study:

• To review and stratify existing AI/ML approaches for PK/PD analysis.
• To position AI/ML methods within the broader scientific computing landscape for PK/PD.
• To examine how AI/ML can contribute to PK/PD time-course modeling.

Main Methods:

• Literature review of AI/ML applications in PK/PD modeling.
• Categorization of AI/ML techniques based on data availability and research questions.
• Analysis of AI/ML's role in PK/PD time-course modeling.

Main Results:

• AI/ML offers diverse applications in PK/PD modeling, adaptable to different data scenarios.
• Stratification of AI/ML approaches provides a framework for their use in PK/PD analysis.
• The integration of AI/ML enhances the capabilities of computational modeling in pharmacology.

Conclusions:

• AI/ML presents significant opportunities to advance PK/PD time-course modeling.
• Strategic application of AI/ML can optimize drug development and patient care.
• Further exploration and integration of AI/ML are recommended for the pharmacometrics community.