基于BP神经网络算法构建糖尿病早期肾病风险预测模型

doi:10.3969/j.issn.1006-5253.2024.02.004

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (4839 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要目的探讨糖尿病早期肾病的相关危险因素，并基于BP神经网络算法构建其风险预测模型。方法回顾性分析永康某中医院2020年1月至2022年12月收治的1 048例糖尿病患者，其中糖尿病肾病患者115例，占10.97%，并以此分为DKD组（糖尿病肾病组115例）和DM组（糖尿病组933例）。收集患者相关资料，采用倾向性评分匹配（PSM）排除混杂因素后按1∶1最邻近方法进行匹配。以单因素分析中具有统计学意义的指标,运用BP神经网络算法基于相关因素构建预测模型。以平均绝对值误差（MAE）进行模型效能分析，以受试者工作特征曲线（ROC）评估风险预测模型的预测价值，并进行外部验证，采用校准曲线评估模型一致性。结果混杂因素有性别、合并高血压、空腹血糖、尿酸，将建模集按1∶1比例以最邻近方法进行PSM排除混杂因素后，DKD组95例，DM组95例。单因素分析结果提示患者年龄、2型糖尿病、总胆固醇（TC）、尿蛋白排泄率、糖尿病病程、胱抑素C(Cys C）组间差异具有统计学意义（P<0.05）。预测精度从大到小依次为BP神经网络算法、决策树、支持向量机、逻辑回归。BP神经网络结果显示影响糖尿病早期肾病发生重要性的前4位因素依次为蛋白尿排泄率、年龄、糖尿病病程、Cys C。预测模型AUC为0.959（95%CI：0.917～1.000），约登指数0.867，对应的敏感度与特异性分别为0.867、1.000。外部验证AUC为0.958（95%CI：0.922～0.995），其敏感度与特异性分别为0.804、1.000，校准图中校准曲线贴近于标准曲线。结论基于机器学习法构建的以年龄、病程、尿蛋白排泄率、TC、Cys C、2型糖尿病为预测特征的BP神经网络算法模型对糖尿病早期肾病有较好的预测价值，可以把该模型临床应用于此类高风险人群的管理识别。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	杜燕华
	朱洪挺

关键词 ： BP神经网络, 糖尿病肾病, 早期肾病, 预测模型, 影响因素

Abstract：Objective To investigate the risk factors of early diabetic nephropathy and construct a risk prediction model based on BP neural network algorithm. Methods A total of 1 048 diabetic patients admitted to Yongkang Hospital of Traditional Chinese Medicine from January 2020 to December 2022 were retrospectively analyzed, including 115 diabetic nephropathy patients (10.97%), and were divided into the DKD group (115 diabetic nephropathy patients) and the DM group (933 diabetic patients). The relevant data of patients were collected and matched according to 1∶1 nearest proximity method after the confounding factors were excluded by propensity score matching (PSM). The prediction model was built based on the correlation factors by using the statistically significant indicators in the single factor analysis and BP neural network algorithm. Mean absolute error (MAE) was used to analyze the model efficacy, the predictive value of the risk prediction model was evaluated by receiver operating characteristic curve (ROC), and external validation was performed. The model consistency was evaluated by calibration curve.Results The confounding factors were gender, hypertension, fasting blood glucose and uric acid. After the modeling set was 1∶1 and PSM was performed by the nearest method, the confounding factors were excluded: 95 cases in DKD group and 95 cases in DM group. Univariate results indicated that there were significant differences in age, type 2 diabetes, total cholesterol, urinary protein excretion rate, diabetes course, and cystatin C(CysC) between groups (P<0.05). The prediction accuracy was BP neural network algorithm, decision tree, support vector machine and logistic regression in the descending order. The results of BP neural network showed that the top 4 factors affecting the occurrence of early diabetic nephropathy were proteinuria excretion rate, age, diabetes course and cystatin C(CysC) in order. The AUC of the prediction model was 0.959 (95%CI: 0.917-1.000), the Yoden index was 0.867, and the corresponding sensitivity and specificity were 0.867 and 1.000, respectively. The external validation-AUC was 0.958 (95%CI: 0.922-0.995), and its sensitivity and specificity were 0.804 and 1.000, respectively. The calibration curve in the calibration diagram was close to the standard curve.Conclusion The BP neural network algorithm model based on machine learning, which takes age, disease course, urinary protein excretion rate, TC, CysC and type 2 diabetes as predictive features, has good predictive value for early diabetic nephropathy, and can be clinically applied to the management and identification of high-risk population.

Key words： back propagation(BP) neural network diabetic nephropathy early kidney disease predictive model influencing factor

收稿日期: 2023-11-06

通讯作者: 杜燕华，Email：158021829@qq.com

引用本文:

杜燕华，朱洪挺. 基于BP神经网络算法构建糖尿病早期肾病风险预测模型[J]. 中国医院统计, 2024, 31(2): 95-101.
Du Yanhua， Zhu Hongting. Risk prediction model for early diabetic kidney disease constructed based on BP neural network algorithm. journal1, 2024, 31(2): 95-101.

链接本文:

http://manu58.magtech.com.cn/Jwk_zgyytj/CN/10.3969/j.issn.1006-5253.2024.02.004 或 http://manu58.magtech.com.cn/Jwk_zgyytj/CN/Y2024/V31/I2/95

［1］CARMICHAEL J, FADAVI H, ISHIBASHI F, et al. Advances in screening, early diagnosis and accurate staging of diabetic neuropathy［J］. Front Endocrinol, 2021,12:671257. DOI: 10.3389/fendo.2021.671257.

［2］TAN S K, PINZON-CORTES J A, COOPER M E. Novel pharmacological interventions for diabetic kidney disease［J］. Curr Opin Nephrol Hypertens, 2024,33(1):13-25. DOI: 10.1097/MNH.0000000000000935.

［3］刘存蓉,史晓伟,罗向霞.早期糖尿病肾病的证型分布及其治疗的理论研究［J/OL］.辽宁中医杂志:1-10［2023-11-03］.DOI:https://link.cnki.net/urlid/21.1128.R.20230823.1137.002.

［4］李慧灵,唐晨晨,申雨萍,等.糖尿病肾病发病机制与中西医治疗的研究进展［J］.临床医学工程,2023,30(8):1175-1176.DOI: 10.3969/j.issn.1674-4659.2023.08.1175.

［5］宋媛媛,陈兵,刘静静,等.BP神经网络模型与Logistic回归模型在精神分裂症患者复发影响因素分析中的比较［J］.护理学报,2022,29(13):59-65.DOI: 10.16460/j.issn1008-9969.2022.13.059.

［6］国家老年医学中心,中华医学会老年医学分会,中国老年保健协会糖尿病专业委员会.中国老年糖尿病诊疗指南:2021年版［J］.中华糖尿病杂志,2021,13(1):14-46.DOI: 10.3760/cma.j.cn115791-20201209-00707.

［7］中华医学会糖尿病学分会微血管并发症学组.中国糖尿病肾脏疾病防治临床指南［J］.中华糖尿病杂志,2019,11(1):15-28.DOI: 10.3760/cma.j.issn.1674-5809.2019.01.004.

［8］WU B T, AN N, YANG Y W, et al. Further understanding and paying attention to normoalbuminuric diabetic kidney disease［J］. Chinese Preventive Medicine, 2023,57(10):1663-1673. DOI: 10.3760/cma.j.cn112150-20230404-00256.

［9］DE MYTTENAERE A, GOLDEN B, LE GRAND B, et al. Mean Absolute Percentage Error for regression models［J］. Neurocomputing, 2016,192:38-48. DOI: 10.1016/j.neucom.2015.12.114.

［10］健康中国行动推进委员会.健康中国行动(2019—2030年):总体要求、重大行动及主要指标［J］.中国循环杂志,2019,34(9):846-858.DOI: 10.3969/j.issn.1000-3614.2019.09.003.

［11］王旭,汪楠,吴国仲,等.阿托伐他汀对合并有糖尿病的终末期肾病病人动静脉内瘘功能的影响研究［J］.安徽医药,2023,27(9):1877-1881.DOI: 10.3969/j.issn.1009-6469.2023.09.041.

［12］马超,郑云鹏.五苓散加减与盐酸贝那普利片联用治疗老年糖尿病肾病的效果［J］.中国医学创新,2023,20(23):32-36.DOI: 10.3969/j.issn.1674-4985.2023.23.008.

［13］常杰.血清视黄醇结合蛋白与尿微量白蛋白在2型糖尿病早期肾病中的诊断价值探讨［J］.中国全科医学,2020,23(S2):132-133.

［14］GUO J, ZHENG H J, ZHANG W T, et al. Accelerated kidney aging in diabetes mellitus［J］. Oxid Med Cell Longev, 2020,2020:1234059. DOI: 10.1155/2020/1234059.

［15］陈怡,詹明.衰老与老年糖尿病肾病的发病机制研究进展［J］.现代实用医学,2023,35(5):694-696.DOI: 10.3969/j.issn.1671-0800.2023.05.043.

［16］余姝,贾珏,王涛涛,等.慢性肾脏病管理对糖尿病肾病进展及透析启动的影响［J］.中国血液净化,2023,22(6):415-420.DOI: 10.3969/j.issn.1671-4091.2023.06.004.

［17］吕以培,符春晖,张素华,等.糖尿病前期、糖尿病正常尿白蛋白与糖尿病早期肾病者的内皮功能障碍与胰岛素抵抗［J］.中国现代医学杂志,2007,17(17):2102-2106.DOI: 10.3969/j.issn.1005-8982.2007.17.015.

［18］王影,周明俊,朱倩文,等.HIF-1α活化介导的胆固醇稳态失调加速大鼠的糖尿病肾病进展［J］.南方医科大学学报,2023,9(10):1782-1788.

［19］HU Y L, DU X G. Blood lipid indicators and different clinical classifications of dyslipidemia and diabetic kidney disease: Correlation and predictive value［J］. Journal of Sichuan University(Medical Sciences), 2023,54(5):1013-1018. DOI: 10.12182/20230960103.

［20］朱珠,耿晨,赵雪飘.胱抑素C、糖化血红蛋白及其交互作用与孕妇妊娠期糖尿病的关系［J］.中国妇幼健康研究,2023,34(10):40-46.DOI: 10.3969/j.issn.1673-5293.2023.10.007.

［21］姜留,周云,张百中,等.糖尿病肾病患者25(OH)D、Hcy、胱抑素C水平变化及意义［J］.中国卫生工程学,2023,22(5):692-694.DOI: 10.19937/j.issn.1671-4199.2023.05.039.

［22］陈粲,刘礼斌.糖尿病肾病患者并发泌尿系感染的logistic分析及血清CRP和Cys-C水平变化［J］.热带医学杂志,2023,23(9):1290-1293.DOI: 10.3969/j.issn.1672-3619.2023.09.022.