Application of artificial neural network in predicting tacrolimus concentrations in kidney transplantation recipients
- Author:
Xiao-Hua FU
1
Author Information
- Publication Type:Journal Article
- Keywords: Artificial neural network; Individualization of dosage regimen; Kidney transplantation; Tacrolimus
- From: Chinese Pharmaceutical Journal 2013;48(12):1000-1004
- CountryChina
- Language:Chinese
- Abstract: OBJECTIVE: To establish an artificial neural network (ANN) for predicting tacrolimus concentrations in kidney transplantation recipients. METHODS: Three hundred and twenty-two tacrolimus concentration data from 55 Chinese kidney transplantation recipients were collected. ANN was established after the network parameters were optimized by using momentum method combined with genetic algorithm. Furthermore, the performance of ANN was compared with that of multiple linear regression (MLR). RESULTS: When using the accumulated dose of tacrolimus in the 6 d before TDM as the input factor, the mean prediction error and mean absolute prediction error of ANN were (0.13±1.91) and (1.49±1.22) ng·L-1, respectively. The absolute prediction errors for 87.9% of the test data set were less than 3.0 ng·L-1. The accuracy and precision of ANN were superior to those of MLR. CONCLUSION: The correlation, accuracy and precision of ANN are good enough to predict tacrolimus concentration.
