A Deep-Learning-based Two-Compartment Predictive Model (PKRNN-2CM) for Vancomycin Therapeutic Drug Monitoring

This study developed a two-compartment deep learning model (PKRNN-2CM) for therapeutic drug monitoring (TDM) of vancomycin (VAN), a commonly used antibiotic. The model, which uses irregularly sampled electronic health record (EHR) data, outperformed a one-compartment model (PKRNN) in predicting VAN concentration. Simulation results also demonstrated the superiority of the PKRNN-2CM model, suggesting that it could improve the accuracy and effectiveness of personalized VAN TDM, leading to better clinical outcomes.