A Deep Learning-Based Approach for Prediction of Vancomycin Treatment Drug Monitoring: Retrospective Data of Critically Ill Patients.

Researchers

Journal

Modalities

Models

Abstract

Vancomycin pharmacokinetics are highly variable in critically ill patients, and clinicians commonly use population pharmacokinetic (PPK) models based on a Bayesian approach to dose. However, these models are population-dependent and may only sometimes meet the needs of individual patients and are only used by experienced clinicians as a reference for making treatment decisions. To assist real-world clinicians, we developed a deep learning-based decision-making system that predicts vancomycin therapeutic drug monitoring (TDM) levels in ICU patients.The goal of our study was to establish JointMLP, a new deep-learning model for predicting vancomycin TDM levels, and compare its performance with the PPK model, XGBoost, and TabNet.We used a 977-case data set split into training and testing groups in a 9:1 ratio. We performed external validation of the model using 1,429 cases from KNUH and 2,394 cases from MIMIC-IV. In addition, we performed 10-fold cross-validation on the internal training dataset and calculated the 95% confidence intervals using the metric. Finally, we evaluated the generalization ability of the JointMLP model using the MIMIC-IV dataset.Our JointMLP model outperformed other models in predicting vancomycin TDM levels in internal and external datasets. Compared to PPK, the JointMLP model improved predictive power by up to 31% (MAE: 6.68 vs. 5.11) on the internal data set and 81% (MAE: 11.87 vs. 6.56) on the external data set. In addition, the JointMLP model significantly outperforms XGBoost and TabNet, with a 13% (MAE: 5.75 vs. 5.11) and 14% (MAE: 5.85 vs. 5.11) improvement in predictive accuracy on the inner dataset, respectively. On both the internal and external datasets, our JointMLP model performed well compared to XGBoost and TabNet, achieving prediction accuracy improvements of 34% and 14%, respectively. Additionally, our JointMLP model showed higher robustness to outlier data than the other models, as evidenced by its higher RMSE performance across all data sets. The mean errors and variances of the JointMLP model were close to zero and smaller than those of the PPK model in internal and external data sets.Our JointMLP approach can help optimize treatment outcomes in critically ill patients in an ICU setting, reducing side effects associated with suboptimal vancomycin administration. These include increased risk of bacterial resistance, extended hospital stays, and increased health care costs. In addition, the superior performance of our model compared to existing models highlights its potential to help real-world clinicians.