Comparison of Machine Learning Algorithms and Nomogram Construction for Diabetic Retinopathy Prediction in Type 2 Diabetes Mellitus Patients.

IF 2 4区 医学 Q2 OPHTHALMOLOGY Ophthalmic Research Pub Date : 2024-01-01 Epub Date: 2024-09-04 DOI:10.1159/000541294
Weiliang Jiang, Zijing Li
{"title":"Comparison of Machine Learning Algorithms and Nomogram Construction for Diabetic Retinopathy Prediction in Type 2 Diabetes Mellitus Patients.","authors":"Weiliang Jiang, Zijing Li","doi":"10.1159/000541294","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>The aim of this study was to compare various machine learning algorithms for constructing a diabetic retinopathy (DR) prediction model among type 2 diabetes mellitus (DM) patients and to develop a nomogram based on the best model.</p><p><strong>Methods: </strong>This cross-sectional study included DM patients receiving routine DR screening. Patients were randomly divided into training (244) and validation (105) sets. Least absolute shrinkage and selection operator regression was used for the selection of clinical characteristics. Six machine learning algorithms were compared: decision tree (DT), k-nearest neighbours (KNN), logistic regression model (LM), random forest (RF), support vector machine (SVM), and XGBoost (XGB). Model performance was assessed via receiver-operating characteristic (ROC), calibration, and decision curve analyses (DCAs). A nomogram was then developed on the basis of the best model.</p><p><strong>Results: </strong>Compared with the five other machine learning algorithms (DT, KNN, RF, SVM, and XGB), the LM demonstrated the highest area under the ROC curve (AUC, 0.894) and recall (0.92) in the validation set. Additionally, the calibration curves and DCA results were relatively favourable. Disease duration, DPN, insulin dosage, urinary protein, and ALB were included in the LM. The nomogram exhibited robust discrimination (AUC: 0.856 in the training set and 0.868 in the validation set), calibration, and clinical applicability across the two datasets after 1,000 bootstraps.</p><p><strong>Conclusion: </strong>Among the six different machine learning algorithms, the LM algorithm demonstrated the best performance. A logistic regression-based nomogram for predicting DR in type 2 DM patients was established. This nomogram may serve as a valuable tool for DR detection, facilitating timely treatment.</p>","PeriodicalId":19662,"journal":{"name":"Ophthalmic Research","volume":null,"pages":null},"PeriodicalIF":2.0000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ophthalmic Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1159/000541294","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/9/4 0:00:00","PubModel":"Epub","JCR":"Q2","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Introduction: The aim of this study was to compare various machine learning algorithms for constructing a diabetic retinopathy (DR) prediction model among type 2 diabetes mellitus (DM) patients and to develop a nomogram based on the best model.

Methods: This cross-sectional study included DM patients receiving routine DR screening. Patients were randomly divided into training (244) and validation (105) sets. Least absolute shrinkage and selection operator regression was used for the selection of clinical characteristics. Six machine learning algorithms were compared: decision tree (DT), k-nearest neighbours (KNN), logistic regression model (LM), random forest (RF), support vector machine (SVM), and XGBoost (XGB). Model performance was assessed via receiver-operating characteristic (ROC), calibration, and decision curve analyses (DCAs). A nomogram was then developed on the basis of the best model.

Results: Compared with the five other machine learning algorithms (DT, KNN, RF, SVM, and XGB), the LM demonstrated the highest area under the ROC curve (AUC, 0.894) and recall (0.92) in the validation set. Additionally, the calibration curves and DCA results were relatively favourable. Disease duration, DPN, insulin dosage, urinary protein, and ALB were included in the LM. The nomogram exhibited robust discrimination (AUC: 0.856 in the training set and 0.868 in the validation set), calibration, and clinical applicability across the two datasets after 1,000 bootstraps.

Conclusion: Among the six different machine learning algorithms, the LM algorithm demonstrated the best performance. A logistic regression-based nomogram for predicting DR in type 2 DM patients was established. This nomogram may serve as a valuable tool for DR detection, facilitating timely treatment.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
比较机器学习算法和用于 2 型糖尿病患者糖尿病视网膜病变预测的示意图构建。
简介:目的比较用于构建 2 型糖尿病(DM)患者糖尿病视网膜病变(DR)预测模型的各种机器学习算法,并根据最佳模型制定提名图:这项横断面研究包括接受常规糖尿病视网膜病变筛查的糖尿病患者。患者被随机分为训练集(244 例)和验证集(105 例)。临床特征选择采用最小绝对收缩和选择算子回归。比较了六种机器学习算法:决策树(DT)、k-近邻(KNN)、逻辑回归模型(LM)、随机森林(RF)、支持向量机(SVM)和 XGBoost(XGB)。模型性能通过接收器操作特征(ROC)、校准和决策曲线分析(DCA)进行评估。然后在最佳模型的基础上建立了一个提名图:与其他五种机器学习算法(DT、KNN、RF、SVM 和 XGB)相比,LM 在验证集中的 ROC 曲线下面积(AUC,0.894)和召回率(0.92)最高。此外,校准曲线和 DCA 结果也相对较好。病程、DPN、胰岛素用量、尿蛋白和 ALB 均被纳入 LM。经过 1,000 次引导后,该提名图在两个数据集上显示出强大的辨别能力(训练集的 AUC 为 0.856,验证集的 AUC 为 0.868)、校准能力和临床适用性:在六种不同的机器学习算法中,LM 算法表现最佳。结论:在六种不同的机器学习算法中,LM 算法的性能最佳,它建立了一个基于逻辑回归的提名图,用于预测 2 型糖尿病患者的 DR。该提名图可作为检测 DR 的重要工具,促进及时治疗。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Ophthalmic Research
Ophthalmic Research 医学-眼科学
CiteScore
3.80
自引率
4.80%
发文量
75
审稿时长
6-12 weeks
期刊介绍: ''Ophthalmic Research'' features original papers and reviews reporting on translational and clinical studies. Authors from throughout the world cover research topics on every field in connection with physical, physiologic, pharmacological, biochemical and molecular biological aspects of ophthalmology. This journal also aims to provide a record of international clinical research for both researchers and clinicians in ophthalmology. Finally, the transfer of information from fundamental research to clinical research and clinical practice is particularly welcome.
期刊最新文献
Anp32b-deficiency suppresses ocular development by repression of Pax6. Analysis of retinal and choroidal microvasculature in systemic sclerosis using optical coherence tomography angiography: A systematic review and meta-analysis. Longitudinal changes in choroidal thickness and choroidal vascularity index in age-related macular degeneration. Efficacy of intracameral mydriatics in pediatric lens surgery. Intraocular Pressure Fluctuation as Assessed by Water Drinking Test Following iStent inject Implantation in Open Angle Glaucoma Patients.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1