Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning.

IF 2.2 4区医学 Q2 UROLOGY & NEPHROLOGY

BMC Nephrology Pub Date : 2024-10-19 DOI:10.1186/s12882-024-03809-2

John W Larkin, Suman Lama, Sheetal Chaudhuri, Joanna Willetts, Anke C Winter, Yue Jiao, Manuela Stauss-Grabo, Len A Usvyat, Jeffrey L Hymes, Franklin W Maddux, David C Wheeler, Peter Stenvinkel, Jürgen Floege

{"title":"Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning.","authors":"John W Larkin, Suman Lama, Sheetal Chaudhuri, Joanna Willetts, Anke C Winter, Yue Jiao, Manuela Stauss-Grabo, Len A Usvyat, Jeffrey L Hymes, Franklin W Maddux, David C Wheeler, Peter Stenvinkel, Jürgen Floege","doi":"10.1186/s12882-024-03809-2","DOIUrl":null,"url":null,"abstract":"Background: Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient's 180-day GIB hospitalization risk.Methods: An eXtreme Gradient Boosting (XGBoost) and logistic regression model were developed using an HD dataset in United States (2017-2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments ≤ 180 days before GIB hospitalization were classified as positive observations; others were negative. Models considered 1,303 exposures/covariates. Performance was measured using unseen testing data.Results: Incidence of 180-day GIB hospitalization was 1.18% in HD population (n = 451,579), and 1.12% in testing dataset (n = 38,853). XGBoost showed area under the receiver operating curve (AUROC) = 0.74 (95% confidence interval (CI) 0.72, 0.76) versus logistic regression showed AUROC = 0.68 (95% CI 0.66, 0.71). Sensitivity and specificity were 65.3% (60.9, 69.7) and 68.0% (67.6, 68.5) for XGBoost versus 68.9% (64.7, 73.0) and 57.0% (56.5, 57.5) for logistic regression, respectively. Associations in exposures were consistent for many factors. Both models showed GIB hospitalization risk was associated with older age, disturbances in anemia/iron indices, recent all-cause hospitalizations, and bone mineral metabolism markers. XGBoost showed high importance on outcome prediction for serum 25 hydroxy (25OH) vitamin D levels, while logistic regression showed high importance for parathyroid hormone (PTH) levels.Conclusions: Machine learning can be considered for early detection of GIB event risk in HD. XGBoost outperforms logistic regression, yet both appear suitable. External and prospective validation of these models is needed. Association between bone mineral metabolism markers and GIB events was unexpected and warrants investigation.Trial registration: This retrospective analysis of real-world data was not a prospective clinical trial and registration is not applicable.","PeriodicalId":9089,"journal":{"name":"BMC Nephrology","volume":"25 1","pages":"366"},"PeriodicalIF":2.2000,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11490046/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BMC Nephrology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s12882-024-03809-2","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"UROLOGY & NEPHROLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient's 180-day GIB hospitalization risk.

Methods: An eXtreme Gradient Boosting (XGBoost) and logistic regression model were developed using an HD dataset in United States (2017-2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments ≤ 180 days before GIB hospitalization were classified as positive observations; others were negative. Models considered 1,303 exposures/covariates. Performance was measured using unseen testing data.

Results: Incidence of 180-day GIB hospitalization was 1.18% in HD population (n = 451,579), and 1.12% in testing dataset (n = 38,853). XGBoost showed area under the receiver operating curve (AUROC) = 0.74 (95% confidence interval (CI) 0.72, 0.76) versus logistic regression showed AUROC = 0.68 (95% CI 0.66, 0.71). Sensitivity and specificity were 65.3% (60.9, 69.7) and 68.0% (67.6, 68.5) for XGBoost versus 68.9% (64.7, 73.0) and 57.0% (56.5, 57.5) for logistic regression, respectively. Associations in exposures were consistent for many factors. Both models showed GIB hospitalization risk was associated with older age, disturbances in anemia/iron indices, recent all-cause hospitalizations, and bone mineral metabolism markers. XGBoost showed high importance on outcome prediction for serum 25 hydroxy (25OH) vitamin D levels, while logistic regression showed high importance for parathyroid hormone (PTH) levels.

Conclusions: Machine learning can be considered for early detection of GIB event risk in HD. XGBoost outperforms logistic regression, yet both appear suitable. External and prospective validation of these models is needed. Association between bone mineral metabolism markers and GIB events was unexpected and warrants investigation.

Trial registration: This retrospective analysis of real-world data was not a prospective clinical trial and registration is not applicable.

查看原文本刊更多论文

利用机器学习预测血液透析中消化道出血的住院风险。

背景：胃肠道出血（GIB）是肾衰竭的临床难题。INSPIRE 小组评估了机器学习能否确定血液透析（HD）患者 180 天 GIB 住院风险：使用美国（2017-2020 年）的 HD 数据集开发了一个梯度提升（XGBoost）和逻辑回归模型。患者数据被随机分割（50%为训练数据，30%为验证数据，20%为测试数据）。GIB住院前≤180天的HD治疗被归类为阳性观察结果；其他为阴性观察结果。模型考虑了 1,303 个暴露/变量。使用未见过的测试数据来衡量性能：在 HD 人口（n = 451,579 人）中，180 天 GIB 住院率为 1.18%，在测试数据集（n = 38,853 人）中为 1.12%。XGBoost 的接收器工作曲线下面积 (AUROC) = 0.74（95% 置信区间 (CI) 0.72，0.76），而逻辑回归的接收器工作曲线下面积 (AUROC) = 0.68（95% 置信区间 (CI) 0.66，0.71）。XGBoost 的灵敏度和特异性分别为 65.3% (60.9, 69.7) 和 68.0% (67.6, 68.5)，而逻辑回归的灵敏度和特异性分别为 68.9% (64.7, 73.0) 和 57.0% (56.5, 57.5)。许多因素的暴露相关性是一致的。两个模型均显示，GIB 住院风险与年龄偏大、贫血/铁指数紊乱、近期全因住院和骨矿物质代谢指标有关。XGBoost显示血清25羟基（25OH）维生素D水平对结果预测的重要性很高，而逻辑回归显示甲状旁腺激素（PTH）水平的重要性很高：结论：机器学习可用于早期检测血液透析患者的 GIB 事件风险。XGBoost优于逻辑回归，但两者似乎都适用。需要对这些模型进行外部和前瞻性验证。骨矿物质代谢标志物与GIB事件之间的关联出乎意料，值得研究：这项对真实世界数据的回顾性分析并非前瞻性临床试验，因此注册并不适用。

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分钟内获得全文求助全文

来源期刊

BMC Nephrology UROLOGY & NEPHROLOGY-

CiteScore

4.30

自引率

0.00%

发文量

375

审稿时长

3-8 weeks

期刊介绍： BMC Nephrology is an open access journal publishing original peer-reviewed research articles in all aspects of the prevention, diagnosis and management of kidney and associated disorders, as well as related molecular genetics, pathophysiology, and epidemiology.