Báo cáo y học: A novel approach for prediction of tacrolimus blood concentration in liver transplantation patients in the intensive care unit through support vector regression

Tuyển tập các báo cáo nghiên cứu về y học được đăng trên tạp chí y học Critical Care giúp cho các bạn có thêm kiến thức về ngành y học đề tài: A novel approach for prediction of tacrolimus blood concentration in liver transplantation patients in the intensive care unit through support vector regression...
Nội dung trích xuất từ tài liệu:
Báo cáo y học: " A novel approach for prediction of tacrolimus blood concentration in liver transplantation patients in the intensive care unit through support vector regression" Available online http://ccforum.com/content/11/4/R83Research Open AccessVol 11 No 4A novel approach for prediction of tacrolimus blood concentrationin liver transplantation patients in the intensive care unit throughsupport vector regressionStijn Van Looy1*, Thierry Verplancke2*, Dominique Benoit2, Eric Hoste2, Georges Van Maele3,Filip De Turck1 and Johan Decruyenaere21Ghent University, Department of Information Technology (INTEC), Gaston Crommenlaan 8, Ghent, Belgium2Ghent University Hospital, Intensive Care Department, De Pintelaan 185, Ghent, Belgium3Ghent University, Department of Medical Statistics, De Pintelaan 185, Ghent, Belgium* Contributed equallyCorresponding author: Thierry Verplancke, Thierry.Verplancke@UGent.beReceived: 8 May 2007 Revisions requested: 11 Jul 2007 Revisions received: 23 Jul 2007 Accepted: 26 Jul 2007 Published: 26 Jul 2007Critical Care 2007, 11:R83 (doi:10.1186/cc6081)This article is online at: http://ccforum.com/content/11/4/R83© 2007 Van Looy et al.; licensee BioMed Central Ltd.This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.AbstractIntroduction Tacrolimus is an important immunosuppressive data values. Nonlinear RBF SVR, linear SVR, and MLR weredrug for organ transplantation patients. It has a narrow performed after selection of clinically relevant input variablestherapeutic range, toxic side effects, and a blood concentration and model parameters. Differences between observed andwith wide intra- and interindividual variability. Hence, it is of the predicted tacrolimus blood concentrations were calculated.utmost importance to monitor tacrolimus blood concentration, Prediction accuracy of the three methods was compared afterthereby ensuring clinical effect and avoiding toxic side effects. fivefold cross-validation (Friedman test and Wilcoxon signedPrediction models for tacrolimus blood concentration can rank analysis).improve clinical care by optimizing monitoring of theseconcentrations, especially in the initial phase after Results Linear SVR and nonlinear RBF SVR had mean absolutetransplantation during intensive care unit (ICU) stay. This is the differences between observed and predicted tacrolimus bloodfirst study in the ICU in which support vector machines, as a new concentrations of 2.31 ng/ml (standard deviation [SD] 2.47)data modeling technique, are investigated and tested in their and 2.38 ng/ml (SD 2.49), respectively. MLR had a meanprediction capabilities of tacrolimus blood concentration. Linear absolute difference of 2.73 ng/ml (SD 3.79). The differencesupport vector regression (SVR) and nonlinear radial basis between linear SVR and MLR was statistically significant (p