[1]郑尔昌,邹金串,薛成斌,等.糖尿病联合并发症发病风险计算与预测[J].华侨大学学报(自然科学版),2022,43(4):498-510.[doi:10.11830/ISSN.1000-5013.202110012]
　ZHENG Erchang,ZOU Jinchuan,XUE Chengbin,et al.Risk Calculation and Prediction of Diabetes Combined Complications Incidence[J].Journal of Huaqiao University(Natural Science),2022,43(4):498-510.[doi:10.11830/ISSN.1000-5013.202110012]
点击复制

糖尿病联合并发症发病风险计算与预测()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第43卷
期数:
2022年第4期
页码:
498-510
栏目:
出版日期:
2022-07-18

文章信息/Info

Title:
Risk Calculation and Prediction of Diabetes Combined Complications Incidence
文章编号:
1000-5013(2022)04-0498-13
作者:
郑尔昌1;  邹金串2;  薛成斌3;  张晋伟1;  陈少阳4;  陈强4;  胡国鹏1
1. 华侨大学 体育与健康科学研究中心, 福建 泉州, 362021;2. 华侨大学 旅游学院, 福建 泉州, 362021;3. 仰恩大学 管理学院, 福建 泉州, 362014;4. 福建省泉州市丰泽区华大街道社区卫生服务中心, 福建 泉州, 362021
Author(s):
ZHENG Erchang1;  ZOU Jinchuan2;  XUE Chengbin3;  ZHANG Jinwei1; CHEN Shaoyang4;  CHEN Qiang4;  HU Guopeng1
1. Sports and Health Science Research Center, Huaqiao University, Quanzhou 362021, China; 2. College of Tourism, Huaqiao University, Quanzhou 362021, China; 3. College of Management, Yang’en University, Quanzhou 362014, China; 4. Community Health Service Center of Huada Street of Quanzhou Fengze District of Fujian Province, Quanzhou 362021, China
关键词:
糖尿病;  并发症;  关联发病率;  关键因素;  发病预测;  关联规则;  随机森林
Keywords:
diabetes;  complications;  related incidence rate;  key factor;  incidence prediction;  association rules;  random forest
分类号:
R587.1;TP181
DOI:
10.11830/ISSN.1000-5013.202110012
文献标志码:
A
摘要:
采用国家人口与健康科学数据共享平台临床医学科学数据中心提供的3 000例糖尿病并发症数据作为数据集,对糖尿病联合并发症发病风险进行计算与预测.通过关联规则查找高风险联合并发症并计算各联合并发症的关联发病率,采用随机森林算法建立高风险联合并发症发病预测模型,并查找其关键影响因素.研究结果表明:部分联合并发症关联发病率超过90%;在筛选出的12组高风险联合并发症中,高血压、动脉粥样硬化、视网膜病变、冠心病、肾病等是常见并发症;不同的联合并发症中关键影响因素(生化指标)各不相同;各联合并发症十折交叉验证法的分类平均精度均在0.800 0以上,曲线下面积(AUC)值均大于0.67.
Abstract:
Using the data of 3 000 cases of diabetes complications provided by Clinical Medical Science Center of the National Population and Health Science Data Sharing Platform as a data set, the risk of diabetes combined complications was calculated and predicted. High-risk combined complications were found through association rules and the associated morbidity rate of each combined complication was calculated. The random forest algorithm was used to establish a high-risk combined complication incidence prediction model, and its key influencing factors were found. The research results show that: the related incidence rate of partial combined complications exceeds 90%. Among the slected 12 groups of high-risk combined complications, hypertension, atherosclerosis, retinopathy, coronary heart disease, nephropathy, et al are common complications. The key influencing factors(biochemical indices)in different combined combinations are different. The classification average accuracy of each combined ten fold cross validation method is over 0.800 0, and the area under curve(AUC)value is all greater than 0.67.

参考文献/References:

[1] 张争辉,薛爱芹,于兰.糖尿病相关研究进展[J].世界最新医学信息文摘,2019,19(20):145,149.DOI:10.19613/j.cnki.1671-3141.2019.20.067.
[2] TABAEI B P,HERMAN W H.A multivariate logistic regression equation to screen for diabetes: Development and validation[J].Diabetes Care,2002,25(11):1999-2003.DOI:10.2337/diacare.25.11.1999.
[3] FATIMA M,PASHA M.Survey of machine learning algorithms for disease diagnostic[J].Journal of Intelligent Learning Systems and Applications,2017,9(1):1-16.DOI:10.4236/jilsa.2017.91001.
[4] SOWJANYA K,SINGHAL A,CHOUDHARY C.MobDBTest: A machine learning based system for predicting diabetes risk using mobile devices[C]//IEEE International Advance Computing Conference.Banglore:IEEE Press,2015:397-402.DOI:10.1109/IADCC.2015.7154738.
[5] 谭昭,李文歌.2型糖尿病患者血清尿酸及尿微量白蛋白水平与慢性血管并发症的相关性[J].中国医科大学学报,2018,47(1):67-72.DOI:10.12007/j.issn.0258-4646.2018.01.015.
[6] 李晓燕,孟凡杰,段玉龙,等.改良早期预警评分、血糖值评分及两评分结合预测糖尿病急性并发症患者预后能力的对比研究[J].实用医学杂志,2018,34(3):397-400.DOI:10.3969/j.issn.1006-5725.2018.03.014.
[7] 王雷.经颅多普勒微栓子监测在糖尿病脑血管病中的应用效果及对并发症的预测价值[J].中国医药科学,2020,10(5):201-203,214.DOI:10.3969/j.issn.2095-0616.2020.05.058.
[8] 邢美艳,姜天,夏莉,等.皮肤无创晚期糖基化终末产物测定在社区2型糖尿病血管性并发症筛查中的作用研究[J].中国全科医学,2020,23(8):913-919.DOI:10.12114/j.issn.1007-9572.2020.00.039.
[9] CEDERHOLM J,KATARINA E O,ELIASSON B,et al.Risk prediction of cardiovascular disease in type 2 diabetes: A risk equation from the Swedish National diabetes register[J].Diabetes Care,2008,31(10):2038-2043.DOI:10.2337/dc08-0662.
[10] 张振堂,杨洋,韩福俊,等.基于社区2型糖尿病患者的心脑血管事件5年风险预测模型[J].山东大学学报(医学版),2017,55(6):108-113.DOI:10.6040/j.issn.1671-7554.0.2017.341.
[11] 明淑萍,刘玲,周黎,等.糖尿病急性并发症继发轻度认知功能障碍的预测模型及时间窗分析[J].中风与神经疾病杂志,2017,34(9):786-791.DOI:10.19845/j.cnki.zfysjjbzz.2017.09.004.
[12] 王洁,乔艺璇,彭岩,等.基于Logistic回归和多层神经网络的Ⅱ型糖尿病并发症预测[J].高技术通讯,2019,29(5):455-461.DOI:10.3772/j.issn.1002-0470.2019.05.006.
[13] 王冰蓉,孙阳,李益颖,等.糖化血红蛋白联合非传统血糖监测指标对妊娠期糖尿病患者急慢性并发症的预测价值[J].创伤与急诊电子杂志,2019,7(1):22-28.DOI:10.16746/j.cnki.11-9332/r.2019.01.005.
[14] 徐晓,张蕾,王珍.基于模糊综合评价法的脑中风风险预测系统[J].计算机仿真,2015,32(7):344-347,360.DOI:10.3969/j.issn.1006-9348.2015.07.076.
[15] 李攀.基于神经网络的2型糖尿病并发症预测模型的研究[D].广州:中医药大学,2016.
[16] 崔纯纯.基于神经网络的糖尿病并发症预测系统研究[D].北京:北京交通大学,2018.
[17] VIJIYAKUMAR K, LAVANYA B, NIRMALA I,et al.Random forest algorithm for the prediction of diabetes[C]//IEEE International Conference on System Computation,Automation and Networking.Pondicherry:IEEE Press,2019:1-5.DOI:10.1109/ICSCAN.2019.8878802.
[18] 邱云飞,郭蕾.面向非均衡数据的糖尿病并发症预测[J].数据分析与知识发现,2021,5(2):13.DOI:10.11925/infotech.2096-3467.2020.0353.
[19] HAN Jiawei,KAMBER M,PEI Jian.Data mining: Concept and techniques[M].Beijing: China Machine Press,2012.
[20] 陈纪林.防治动脉粥样硬化的新动向[J].中国循环杂志,2001,16(3):163.

备注/Memo

备注/Memo:
收稿日期: 2021-10-08
通信作者: 胡国鹏(1978-),男,教授,博士,主要从事运动与健康、运动氧科学的研究.E-mail:hugp@hqu.edu.cn.
基金项目: 国家体育总局科研基金资助项目(2017C106); 福建省自然科学基金规划项目(2020J01087); 福建省泉州市丰泽区科技计划项目(2020FZ34)http://www.hdxb.hqu.edu.cn
更新日期/Last Update: 2022-07-20