基于多指标的糖尿病患者视网膜病变预测评估模型研究

doi:10.3969/j.issn.1006-5253.2023.06.001

摘要
图/表
参考文献
相关文章 (0)

全文: PDF (4218 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要目的基于糖尿病患者数据仓储中的疾病及生理生化指标数据，统计分析并进行疾病预测研究。方法通过χ²检验、Cochran-Armitage趋势检验、非参数检验对糖尿病患者是否并发视网膜病变在人口学特征、慢病共患情况、生理生化指标等单因素的分布差异进行统计推断。采用Fisher判别筛选糖尿病并发视网膜病变的价值指标。结果低年龄组糖尿病患者，患有高血压、脑卒中、脂肪肝、肾病、肾衰、下肢动脉病变、血液病的糖尿病患者视网膜病变的并发率高，差异具有统计学意义（P＜0.05）。糖尿病患者视网膜病变的并发率又随着共患疾病种类的增加而增高的趋势（χ²=50.527，P<0.001）。肾病、肾衰、血液病、下肢动脉病变、其他内分泌疾病以及多病共患是糖尿病并发视网膜病变的不利因素。尿微量蛋白、糖化血红蛋白指标对糖尿病患者并发视网膜病变的预判有重要价值。结论糖尿病患者是否并发视网膜病变在患者的患病及生理生化指标间存在统计学差异；通过糖尿病患者的慢性病共患情况、生理生化指标数据能够较好地预测是否并发视网膜病变。

	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
	张娟娟
	赵燕
	许旺
	邱方远
	许建强

关键词 ：糖尿病视网膜病变, 疾病预测, Fisher判别分析

Abstract：Objective To make a statistical analysis and study on disease prediction based on the data of disease, physiological and biochemical indicators in the data warehouse of diabetic patients.Methods Through the chi-square test, Cochran-Armitage trends, and nonparametric test, the differences in demographic characteristics, chronic disease comorbidity, physiological and biochemical indexes distribution of diabetic retinopathy were statistically inferred. Fisher discriminant analysis was used to screen the value of diabetic retinopathy.Results In diabetic patients in the lower age group, the coincidence of retinopathy was higher in diabetic patients with hypertension, stroke, fatty liver, kidney disease, kidney failure, lower limb artery disease, and blood disease, and the difference was statistically significant (P＜0.05). The coincidence of retinopathy in diabetic patients increased with the increase of the types of common diseases (χ²=50.527, P<0.001). Nephropathy, renal failure, hematopathy, lower extremity artery disease, other endocrine diseases and multipathic comorbidities were adverse factors for diabetic retinopathy. Urinary microprotein and glycosylated hemoglobin indexes were important for predicting retinopathy in diabetic patients.Conclusion There are statistical differences in the disease, physiological and biochemical indexes of diabetic patients with retinopathy. By the data of chronic disease comorbidities, physiological and biochemical indexes of diabetic patients can better predict whether retinopathy is complicated or not.

Key words： diabetic retinopathy disease prediction Fisher discriminant analysis

收稿日期: 2023-06-07

基金资助:山东省卫生健康委员会深化医改研究项目（YGY2023056）；
山东省卫生健康委员会全省卫生健康政策研究项目（WZY202383）；
山东省人文社会科学项目（2023-ESDZ-114）；
滨州市社会科学规划项目（23-SKGH-064）

通讯作者: 许旺，Email：1529600042@qq.com

引用本文:

张娟娟，赵燕，许旺，邱方远，许建强. 基于多指标的糖尿病患者视网膜病变预测评估模型研究[J]. 中国医院统计, 2023, 30(6): 401-407.
Zhang Juanjuan，Zhao Yan，Xu Wang，Qiu Fangyuan，Xu Jianqiang. Prediction and evaluation model of diabetic retinopathy based on multiple indicators. journal1, 2023, 30(6): 401-407.

链接本文:

http://manu58.magtech.com.cn/Jwk_zgyytj/CN/10.3969/j.issn.1006-5253.2023.06.001 或 http://manu58.magtech.com.cn/Jwk_zgyytj/CN/Y2023/V30/I6/401

［1］江维维，杨仙荣，冯宇．糖尿病视网膜病变护理研究的计量学分析［J］．护理实践与研究，2023，20(6)：865-869．

［2］张永亮，熊健皓，李明等．糖尿病视网膜病变智能识别的SCSDN泛化性增强方法［J］.中国数字医学，2021，16(7)：82-87.

［3］姜凤伟，刘畅，王影．2型糖尿病泌汗功能与尿微量白蛋白/尿肌酐比值的相关性研究［J］.锦州医科大学学报,2023,44(2):68-73.DOI:10.13847/j.cnki.lnmu.2023.02.020.

［4］田雪晴，游茂．我国医学人工智能在医疗机构发展面临的问题及政策建议［J］．卫生软科学，2019，33(10)：42-44．DOI: 10.3969/j.issn.1003-2800.2019.10.010.

［5］BRISIMI T S, XU T T, WANG T Y, et al. Predicting chronic disease hospitalizations from electronic health records: An interpretable classification approach［J］. Proc IEEE, 2018, 106(4):690-707. DOI: 10.1109/JPROC.2017.2789319.

［6］邓望月，胡大军，鲁碧玉，等．血液透析对糖尿病视网膜病变影响的研究进展［J］．中国血液净化，2023，22(2)：136-140．DOI: 10.3969/j.issn.1671-4091.2023.02.013.

［7］苏婉雯，徐剑容，宋新．空腹血糖、餐后2h血糖结合糖化血红蛋白对于糖尿病视网膜病变风险价值评估［J］．中国处方药，2022，20(6)：104-106．DOI: 10.3969/j.issn.1671-945X.2022.06.043.

［8］DHATARIYA K, HUMBERSTONE A, HASNAT A, et al. The association between mean glycated haemoglobin or glycaemic variability and the development of retinopathy in people with diabetes: A retrospective observational cohort study［J］. Diabetes Ther, 2021, 12(10):2755-2766. DOI: 10.1007/s13300-021-01146-3.

［9］意大利研究出AI诊断新算法：或可提前10年查出老年痴呆症［J］．医学信息学杂志，2017，38(9)：94．

［10］KRITTANAWONG C, ZHANG H J, WANG Z, et al. Artificial intelligence in precision cardiovascular medicine［J］. J Am Coll Cardiol, 2017, 69(21):2657-2664. DOI: 10.1016/j.jacc.2017.03.571.