Predictive value of gradient boosting decision trees for postoperative atelectasis complications in patients with pulmonary destruction.

IF 16.4 1区化学 Q1 CHEMISTRY, MULTIDISCIPLINARY

Accounts of Chemical Research Pub Date : 2024-07-15 eCollection Date: 2024-01-01 DOI:10.62347/IEQE3348

Zhongming Tang, Jifu Tang, Wei Liu, Guoqiang Chen, Chenggang Feng, Aiping Zhang

{"title":"Predictive value of gradient boosting decision trees for postoperative atelectasis complications in patients with pulmonary destruction.","authors":"Zhongming Tang, Jifu Tang, Wei Liu, Guoqiang Chen, Chenggang Feng, Aiping Zhang","doi":"10.62347/IEQE3348","DOIUrl":null,"url":null,"abstract":"Objective: To explore the application value of a gradient boosting decision tree (GBDT) in predicting postoperative atelectasis in patients with destroyed lungs.Methods: A total of 170 patients with damaged lungs who underwent surgical treatment in Chest Hospital of Guangxi Zhuang Autonomous Region from January 2021 to May 2023 were retrospectively selected. The patients were divided into a training set (n = 119) and a validation set (n = 51). Both GBDT algorithm model and Logistic regression model for predicting postoperative atelectasis in patients were constructed. The receiver operating characteristic (ROC) curve, calibration curve and decision curve were used to evaluate the prediction efficiency of the model.Results: The GBDT model indicated that the relative importance scores of the four influencing factors were operation time (51.037), intraoperative blood loss (38.657), presence of lung function (9.126) and sputum obstruction (1.180). Multivariate Logistic regression analysis revealed that operation duration and sputum obstruction were significant predictors of postoperative atelectasis among patients with destroyed lungs within the training set (P = 0.048, P = 0.002). The ROC curve analysis showed that the area under the curve (AUC) for GBDT and Logistic model in the training set was 0.795 and 0.763, and their AUCs in the validation set were 0.776 and 0.811. The GBDT model's predictions closely matched the ideal curve, showing a higher net benefit than the reference line.Conclusions: GBDT model is suitable for predicting the incidence of complications in small samples.","PeriodicalId":1,"journal":{"name":"Accounts of Chemical Research","volume":null,"pages":null},"PeriodicalIF":16.4000,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11301507/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Accounts of Chemical Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.62347/IEQE3348","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

Objective: To explore the application value of a gradient boosting decision tree (GBDT) in predicting postoperative atelectasis in patients with destroyed lungs.

Methods: A total of 170 patients with damaged lungs who underwent surgical treatment in Chest Hospital of Guangxi Zhuang Autonomous Region from January 2021 to May 2023 were retrospectively selected. The patients were divided into a training set (n = 119) and a validation set (n = 51). Both GBDT algorithm model and Logistic regression model for predicting postoperative atelectasis in patients were constructed. The receiver operating characteristic (ROC) curve, calibration curve and decision curve were used to evaluate the prediction efficiency of the model.

Results: The GBDT model indicated that the relative importance scores of the four influencing factors were operation time (51.037), intraoperative blood loss (38.657), presence of lung function (9.126) and sputum obstruction (1.180). Multivariate Logistic regression analysis revealed that operation duration and sputum obstruction were significant predictors of postoperative atelectasis among patients with destroyed lungs within the training set (P = 0.048, P = 0.002). The ROC curve analysis showed that the area under the curve (AUC) for GBDT and Logistic model in the training set was 0.795 and 0.763, and their AUCs in the validation set were 0.776 and 0.811. The GBDT model's predictions closely matched the ideal curve, showing a higher net benefit than the reference line.

Conclusions: GBDT model is suitable for predicting the incidence of complications in small samples.

查看原文本刊更多论文

梯度提升决策树对肺毁损患者术后肺不张并发症的预测价值。

目的探讨梯度提升决策树（GBDT）在预测肺损伤患者术后肺不张中的应用价值：回顾性选取 2021 年 1 月至 2023 年 5 月期间在广西壮族自治区胸科医院接受手术治疗的 170 例肺损伤患者。将患者分为训练集（n = 119）和验证集（n = 51）。建立了预测患者术后肺不张的 GBDT 算法模型和 Logistic 回归模型。采用接收者操作特征曲线（ROC）、校准曲线和决策曲线来评估模型的预测效率：GBDT模型显示，四个影响因素的相对重要性得分分别为手术时间（51.037）、术中失血量（38.657）、肺功能（9.126）和痰液阻塞（1.180）。多变量 Logistic 回归分析显示，手术时间和痰液阻塞是训练集中肺毁损患者术后发生肺不张的重要预测因素（P = 0.048，P = 0.002）。ROC 曲线分析显示，GBDT 模型和 Logistic 模型在训练集中的曲线下面积（AUC）分别为 0.795 和 0.763，在验证集中的曲线下面积（AUC）分别为 0.776 和 0.811。GBDT 模型的预测结果与理想曲线非常吻合，显示出比参考线更高的净效益：GBDT模型适用于预测小样本并发症的发生率。

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

Accounts of Chemical Research 化学-化学综合

CiteScore

31.40

自引率

1.10%

发文量

312

审稿时长

2 months

期刊介绍： Accounts of Chemical Research presents short, concise and critical articles offering easy-to-read overviews of basic research and applications in all areas of chemistry and biochemistry. These short reviews focus on research from the author’s own laboratory and are designed to teach the reader about a research project. In addition, Accounts of Chemical Research publishes commentaries that give an informed opinion on a current research problem. Special Issues online are devoted to a single topic of unusual activity and significance. Accounts of Chemical Research replaces the traditional article abstract with an article "Conspectus." These entries synopsize the research affording the reader a closer look at the content and significance of an article. Through this provision of a more detailed description of the article contents, the Conspectus enhances the article's discoverability by search engines and the exposure for the research.