Machine learning methods applied to pharmacokinetic modelling of remifentanil in healthy volunteers: a multi-method comparison

This study compared the blood concentrations of remifentanil obtained in a previous clinical investigation with the predicted remifentanil concentrations produced by different pharmacokinetic models: a non-linear mixed effects model created by the software NONMEM; an artificial neural network (ANN) model; a support vector machine (SVM) model; and multi-method ensembles. The ensemble created from the mean of the ANN and the non-linear mixed effects model predictions achieved the smallest error and the highest correlation coefficient. The SVM model produced the highest error and the lowest correlation coefficient. Paired t-tests indicated that there was insufficient evidence that the predicted values of the ANN, SVM and two multi-method ensembles differed from the actual measured values at alpha = 0.05. The ensemble method combining the ANN and non-linear mixed effects model predictions outperformed either method alone. These results indicated a potential advantage of ensembles in improving the accuracy and reducing the variance of pharmacokinetic models.

Population pharmacokinetic and pharmacodynamic models of remifentanil in healthy volunteers using artificial neural network analysis

AIMS

An ordinary sigmoid E(max) model could not predict overshoot of electroencephalographic approximate entropy (ApEn) during recovery from remifentanil effect in our previous study. The aim of this study was to evaluate the ability of an artificial neural network (ANN) to predict ApEn overshoot and to evaluate the predictive performance of the pharmacokinetic model, and pharmacodynamic models of ANN with respect to data used.

METHODS

Using a reduced number of ApEn instances (n = 1581) to make NONMEM modelling feasible and complete ApEn data (n = 24 509), the presence of overshoot was assessed. A total of 1077 measured remifentanil concentrations and ApEn data, and a total of 24 509 predicted concentrations and ApEn data were used in the pharmacodynamic model A and B of ANN, respectively. The testing subset of model B (n = 7352) was used to evaluate the ability of ANN to predict overshoot of ApEn. Mean squared error (MSE) was calculated to evaluate the predictive performance of the ANN models.

RESULTS

With complete ApEn data, ApEn overshoot was observed in 66.7% of subjects, but only in 37% with a reduced number of ApEn instances. The ANN model B predicted 77.8% of ApEn overshoot. MSE (95% confidence interval) was 57.1 (3.22, 71.03) for the pharmacokinetic model, 0.148 (0.004, 0.007) for model A and 0.0018 (0.0017, 0.0019) for model B.

CONCLUSIONS

The reduced ApEn instances interfered with the approximation of true electroencephalographic response. ANN predicted 77.8% of ApEn overshoot. The predictive performance of model B was significantly better than that of model A.