Construction and Interpretation of Prediction Model of Teicoplanin Trough Concentration via Machine Learning

Abstract

ObjectiveTo establish an optimal model to predict the teicoplanin trough concentrations by machine learning, and explain the feature importance in the prediction model using the SHapley Additive exPlanation (SHAP) method.MethodsA retrospective study was performed on 279 therapeutic drug monitoring (TDM) measurements obtained from 192 patients who were treated with teicoplanin intravenously at the First Affiliated Hospital of Army Medical University from November 2017 to July 2021. This study included 27 variables, and the teicoplanin trough concentrations were considered as the target variable. The whole dataset was divided into a training group and testing group at the ratio of 8:2, and predictive performance was compared among six different algorithms. Algorithms with higher model performance (top 3) were selected to establish the ensemble prediction model and SHAP was employed to interpret the model.ResultsThree algorithms (SVR, GBRT, and RF) with high R2 scores (0.676, 0.670, and 0.656, respectively) were selected to construct the ensemble model at the ratio of 6:3:1. The model with R2 = 0.720, MAE = 3.628, MSE = 22.571, absolute accuracy of 83.93%, and relative accuracy of 60.71% was obtained, which performed better in model fitting and had better prediction accuracy than any single algorithm. The feature importance and direction of each variable were visually demonstrated by SHAP values, in which teicoplanin administration and renal function were the most important factors.ConclusionWe firstly adopted a machine learning approach to predict the teicoplanin trough concentration, and interpreted the prediction model by the SHAP method, which is of great significance and value for the clinical medication guidance.

Machine learning advances the integration of covariates in population pharmacokinetic models: Valproic acid as an example

“…Compared with the initially proposed XGBoost model, the reduced performance of our simplified XGBoost model indicated the important influences of these features, particularly the blood sampling time and ALB, on the model output. Nevertheless, a 60.00% IR of the simplified optimum XGBoost model on our external dataset suggested its good forecasting performance, considering the prediction accuracy of the predicted TDM within ±30% of the actual TDM in many similar studies that utilized XGBoost models, ranging from 40% to 75% (Huang et al, 2021b;Guo et al, 2021;Zheng et al, 2021;Ma et al, 2022). Based on the simplified optimum XGBoost model, we designed an easy-to-use web application by using only CYP2C19 genotypes and some noninvasive clinical parameters as an MIPD tool for personalized dosing adjustments.…”

Machine Learning: A New Approach for Dose Individualization

“…After removing duplicated articles, 3,346 studies were screened by the title and/or abstract, 3,175 irrelevant studies were excluded and 171 articles were included for full‐text review. Finally, 64 articles related to precision dosing using ML were included for analysis 11–74 . The PRISMA flow diagram representing the study selection process and review results is presented in Figure .…”

Development of Machine Learning-Based Web System for Estimating Pleural Effusion Using Multi-Frequency Bioelectrical Impedance Analyses

“…SHAP (https://pypi.org/ project/shap/0.41.0/ (accessed on 1 July 2023), version 0.41.0) was used to interpret the models. SHAP is calculated based on the constructed LightGBM model and an index to explain the machine learning model based on game theory [26,27].…”