[1]张宇祯,陶砚蕴,谢诚,等.基于遗传规划与进化策略的华法林剂量预测模型*[J].生物医学工程研究,2018,02:182-186.
ZHANG Yuzhen,TAO Yanyun,XIE Cheng,et al.Warfarin dose predictive modeling based on genetic ?programming and evolution strategy[J].Journal of Biomedical Engineering Research,2018,02:182-186.
点击复制
基于遗传规划与进化策略的华法林剂量预测模型*(PDF)
《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]
- 期数:
-
2018年02期
- 页码:
-
182-186
- 栏目:
-
- 出版日期:
-
2018-06-25
文章信息/Info
- Title:
-
Warfarin dose predictive modeling based on genetic ?programming and evolution strategy
- 文章编号:
-
16726278 (2018)020182 05
- 作者:
-
张宇祯1; 陶砚蕴2; 谢诚3; 薛领3; 张其银1; 蒋彬1Δ
-
1. 苏州大学附属第一医院心血管内科,苏州 215006;2.苏州大学 智能结构与系统研究所,苏州 215131;3.苏州大学附属第一医院药学部,苏州 215006
- Author(s):
-
ZHANG Yuzhen1; TAO Yanyun2; XIE Cheng3; XUE Ling3; ZHANG Qiyin1; JIANG Bin1
-
1. Department of Cardiology, the First Affiliated Hospital of Soochow University, Suzhou 215006, China;2. Institute of Intelligent Structure and System, Soochow University, Suzhou 215131;3. Department of Clinical Pharmacology, the First Affiliated Hospital of Soochow University, Suzhou 215006
-
- 关键词:
-
华法林; 精准医疗; 剂量预测模型; 机器学习; 进化算法
- Keywords:
-
Warfarin; Precision medicine; Dose prediction model; Machine learning; Evolutionary algorithm
- 分类号:
-
R318
- DOI:
-
10.19529/j.cnki.1672-6278.2018.02.13
- 文献标识码:
-
A
- 摘要:
-
提出一种基于遗传规划(genetic programming,GP)和进化策略(evolution strategy,ES)的学习方法,命名为遗传规划-进化策略(GPES),建立更准确的华法林剂量预测模型。纳入247例汉族患者。GP进化复杂特征提取,ES进化模型系数,组成模型,得出预测的华法林维持剂量,与线性回归模型、国际华法林药物基因组学联合会模型,及三种机器学习方法相比较。GPES的均方误差(MSE)(1.68×10-2)和预测值在真实值±20%范围内的比例(20%-p)(53.33%)表现最优;其平方相关系数(R2)(69.45%)为次优;GPES在上述3个指标在测试集与训练集中的差值δMSE(0.43×10-2)和 δ20%-p(0.92%)的绝对值最小,δR2(-10.64%)的绝对值为次小。GPES总体表现最优。因此,本研究方法GPES提高了华法林剂量预测模型的趋势相关性、精度、可用性与泛化性。
- Abstract:
-
To develop a method based on genetic programming (GP) and evolution strategy (ES), denoted by GPES, and to improve the accuracy of Warfarin dose predictive model.247 Chinese Han patients were included. Utilizing GP to evolve complex-featured functions, ES evolving model coefficients and random real number of GP function, we generated prediction model. Then we compared our model with a linear regression model, the model developed by the International Warfarin Pharmacogenetics Consortium (IWPC), as well as three machine learning algorithm. Among all the models,GPES got the best mean square error(MSE)(1.68×10-2) and the percentage of patients whose predicted dose of Warfarin were within ±20% of the actual dose (20%-p) (53.33%), the second best squared correlation coefficient (R2) (69.45%). Besides, GPES got the smallest absolute values of the differences between MSE and 20%-p in training set and in test set, and the second smallest absolute values of the differences between R2, respectively, that was δMSE(0.43×10-2), δ20%-p (0.92%) and δR2 (-10.64%).GPES in this study improves the correlation, accuracy, applicability and extensiveness of Warfarin dose predictive model.
备注/Memo
- 备注/Memo:
-
(收稿日期:2017-09-17) 国家自然科学基金资助项目(81700298);苏州市民生科技医疗卫生应用基础研究项目(SYS201736)。△通信作者Email:jbin@suda.edu.cn
更新日期/Last Update:
2018-07-23