摘要
目的 运用人工神经网络技术构建预测模型，实现环孢素A（CsA）的个体化给药设计。方法 收集肾移植患者临床数据，分析CsA血药浓度及给药剂量的影响因素，确定建模因子。采用遗传-反向传播算法(GA-BP)建立两层链式预测模型，构建CsA个体化给药的程序并少量样本试用。结果 初步完成个体化给药的程序编写，10个病例（总共30次）的预测表明，绝大多数次（23/30）的预测给药剂量与实际给药剂量的误差小于15%，超过半数次（17/30）的误差小于10%。结论 可以尝试用人工神经网络进行CsA个体化给药的研究。
Abstract
OBJECTIVE To build a predictive model constructed by artificial neural network， and to adjust administration of cyclosporine A (CsA) individually. METHODS The cases of kidney transplantation were collected. Influence factors on concentration and dosage of cyclosporine A were analyzed and model factors were determined. A scheme of two chained dosage prediction models was established by GA-BP algorithm, and some samples were used to validate the model. RESULTS A tentative program for individual administration was compiled. The predictive doses of cyclosporine A for 10 patients (30 points in all) demonstrated that the deviation of majority of them（23/30）was below 15% and some of them（17/30） was below 10%， compared with actual doses. CONCLUSION An individual administration of cyclosporine A could be attempted by artificial neural network.
关键词
人工神经网络 /
环孢素A /
个体化给药
{{custom_keyword}} /
Key words
artificial neural network /
cyclosporine A /
individual administration
{{custom_keyword}} /
余俊先 史丽敏 王汝龙 李珊 夏杰 程晟 温爱萍 卫红涛.
基于人工神经网络的环孢素A个体化给药设计[J]. 中国药学杂志, 2010, 45(12): 927-930
YU Jun-xin;SHI Li-min;WNG Ru-long;LI Shn;XI Jie;CHENG Sheng;WEN i-ping;WEI Hong-to.
Individual Administration Model for Cyclosporine A Established Using Artificial Neural Network[J]. Chinese Pharmaceutical Journal, 2010, 45(12): 927-930
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LI J H. Methods and Applications of Clinical Therapeutic Drug Monitoring (临床治疗药物监测的方法和应用) [M]. Beijing： People’ Medical Publishing Press， 2003：103-107.
[2] CHEN H， YU J N， XIANG B R. Current new application of artificial neural network (ANN) in clinical pharmacy [J]. Chin J Mod Appl Pharm (中国现代应用药学杂志)， 2002，19(2)： 85-88.
[3] GUAN D L， WANG R L， LV J H， et al. Ciclosporin therapeutic window in Chinese renal transplantation recipient [J].Chin J Nephrol Dialysis Transplantat (肾脏病与透析肾移植杂志)，1994，3(6)： 478-481.
[4] HAN L Q. Artificial Neural Network Tutorial (人工神经网络教程) [M]. Beijing： Beijing University of Post and Telecommunications Press， 2006：128.
[5] JIA Y T， WU J J. Study of population pharmacokinetics of cyclosporine using NONMEM program [J]. Journal of Pediatric Pharmcy (儿科药学杂志)， 2004，10(3)： 65-67.
[6] WU K H， LU W， LI J， et al. Population pharmacokinetic study of cyclosporine and its application [J]. Chin Pharm J (中国药学杂志)， 2006， 41(23)： 1818-1821.
[7] LI H， LI S. A newly FUP-based incremental updating algorithm of association rules [J]. Computer Engineering Science(计算机工程与科学)， 2005， 27(7)： 74-76.
[8] CAMPS G， PORTA B， SORIA. Prediction of cyclosporine dosage in patients after kidney transplantation using neural networks [J]. IEEE Trans Biomed Eng， 2003， 50(4)： 442-448.
[9] MARTIN J D, SORIA E, CAMPS G. Neural networks as ffectivetechniques in clinical management of patients： some case studies [J]. Trans Inst Measurement Control， 2004， 26 (3)： 169-183.
[10] XU C H， AI Y S， CHEN H T. Prediction of plasma level of cyclosporine A in patients after kidney transplantation using neural networks [J]. Chin Hosp Pharm J (中国医院药学杂志)， 2008，28(4)： 276-278.
[11] HU Y H， ZHAI S D， QIU W. Effect of CYP3A5*3 and CYP3A4*18B genetic polymorphisms on cyclosporine pharmacokinetics in renal transplant patients [J]. Chin Pharmacol Bull (中国药理学通报)， 2009，25(3)： 378-382.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(938 KB)
