Semiparametric mixed-effect least squares support vector machine for analyzing pharmacokinetic and pharmacodynamic data

• Published in: Neurocomputing (Amsterdam) 2011-10, Vol.74 (17), p.3412-3419
• Main Authors: Seok, Kyung Ha, Shim, Jooyong, Cho, Daehyeon, Noh, Gyu-Jeong, Hwang, Changha
• Format: Article
• Language: English
• Subjects: Error analysis; Errors; Least squares method; Least squares support vector machine; Mixed-effect model; Pharmacokinetic and pharmacodynamic data; Regression; Regression analysis; Semiparametric model; Statistics; Support vector machines
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2011.05.012

In this paper we propose a semiparametric mixed-effect least squares support vector machine (LS-SVM) regression model for the analysis of pharmacokinetic (PK) and pharmacodynamic (PD) data. We also develop the generalized cross-validation (GCV) method for choosing the hyperparameters which affect the performance of the proposed LS-SVM. The performance of the proposed LS-SVM is compared with those of NONMEM and the regular semiparametric LS-SVM via four measures, which are mean squared error (MSE), mean absolute error (MAE), mean relative absolute error (MRAE) and mean relative prediction error (MRPE). Through paired- t test statistic we find that the absolute values of four measures of the proposed LS-SVM are significantly smaller than those of NONMEM for PK and PD data. We also investigate the coefficient of determinations R 2's of predicted and observed values. The R 2's of NONMEM are 0.66 and 0.59 for PK and PD data, respectively, while the R 2's of the proposed LS-SVM are 0.94 and 0.96. Through cross validation technique we also find that the proposed LS-SVM shows better generalization performance than the regular semiparametric LS-SVM for PK and PD data. These facts indicate that the proposed LS-SVM is an appealing tool for analyzing PK and PD data.