预测成人嗜血细胞淋巴组织细胞增多症 30 天死亡率的实验室数据机器学习。

IF 7.2 2区 医学 Q1 IMMUNOLOGY
Jun Zhou, Mengxiao Xie, Ning Dong, Mingjun Xie, Jingping Liu, Min Wang, Yaman Wang, Hua-Guo Xu
{"title":"预测成人嗜血细胞淋巴组织细胞增多症 30 天死亡率的实验室数据机器学习。","authors":"Jun Zhou, Mengxiao Xie, Ning Dong, Mingjun Xie, Jingping Liu, Min Wang, Yaman Wang, Hua-Guo Xu","doi":"10.1007/s10875-024-01806-6","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Hemophagocytic Lymphohistiocytosis (HLH) carries a high mortality rate. Current existing risk-evaluation methodologies fall short and improved predictive methods are needed. This study aimed to forecast 30-day mortality in adult HLH patients using 11 distinct machine learning (ML) algorithms.</p><p><strong>Methods: </strong>A retrospective analysis on 431 adult HLH patients from January 2015 to September 2021 was conducted. Feature selection was executed using the least absolute shrinkage and selection operator. We employed 11 ML algorithms to create prediction models. The area under the curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, F1 score, calibration curve and decision curve analysis were used to evaluate these models. We assessed feature importance using the SHapley Additive exPlanation (SHAP) approach.</p><p><strong>Results: </strong>Seven independent predictors emerged as the most valuable features. An AUC between 0.65 and 1.00 was noted among the eleven ML algorithms. The gradient boosting decision tree (GBDT) algorithms demonstrated the most optimal performance (1.00 in the training cohort and 0.80 in the validation cohort). By employing the SHAP method, we identified the variables that contributed to the model and their correlation with 30-day mortality. The AUC of the GBDT algorithms was the highest when using the top 4 (ferritin, UREA, age and thrombin time (TT)) features, reaching 0.99 in the training cohort and 0.83 in the validation cohort. Additionally, we developed a web-based calculator to estimate the risk of 30-day mortality.</p><p><strong>Conclusions: </strong>With GBDT algorithms applied to laboratory data, accurate prediction of 30-day mortality is achievable. Integrating these algorithms into clinical practice could potentially improve 30-day outcomes.</p>","PeriodicalId":15531,"journal":{"name":"Journal of Clinical Immunology","volume":null,"pages":null},"PeriodicalIF":7.2000,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine Learning of Laboratory Data in Predicting 30-Day Mortality for Adult Hemophagocytic Lymphohistiocytosis.\",\"authors\":\"Jun Zhou, Mengxiao Xie, Ning Dong, Mingjun Xie, Jingping Liu, Min Wang, Yaman Wang, Hua-Guo Xu\",\"doi\":\"10.1007/s10875-024-01806-6\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Hemophagocytic Lymphohistiocytosis (HLH) carries a high mortality rate. Current existing risk-evaluation methodologies fall short and improved predictive methods are needed. This study aimed to forecast 30-day mortality in adult HLH patients using 11 distinct machine learning (ML) algorithms.</p><p><strong>Methods: </strong>A retrospective analysis on 431 adult HLH patients from January 2015 to September 2021 was conducted. Feature selection was executed using the least absolute shrinkage and selection operator. We employed 11 ML algorithms to create prediction models. The area under the curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, F1 score, calibration curve and decision curve analysis were used to evaluate these models. We assessed feature importance using the SHapley Additive exPlanation (SHAP) approach.</p><p><strong>Results: </strong>Seven independent predictors emerged as the most valuable features. An AUC between 0.65 and 1.00 was noted among the eleven ML algorithms. The gradient boosting decision tree (GBDT) algorithms demonstrated the most optimal performance (1.00 in the training cohort and 0.80 in the validation cohort). By employing the SHAP method, we identified the variables that contributed to the model and their correlation with 30-day mortality. The AUC of the GBDT algorithms was the highest when using the top 4 (ferritin, UREA, age and thrombin time (TT)) features, reaching 0.99 in the training cohort and 0.83 in the validation cohort. Additionally, we developed a web-based calculator to estimate the risk of 30-day mortality.</p><p><strong>Conclusions: </strong>With GBDT algorithms applied to laboratory data, accurate prediction of 30-day mortality is achievable. Integrating these algorithms into clinical practice could potentially improve 30-day outcomes.</p>\",\"PeriodicalId\":15531,\"journal\":{\"name\":\"Journal of Clinical Immunology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":7.2000,\"publicationDate\":\"2024-09-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Clinical Immunology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1007/s10875-024-01806-6\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"IMMUNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Clinical Immunology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s10875-024-01806-6","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"IMMUNOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

背景:嗜血细胞淋巴组织细胞增多症(HLH嗜血细胞淋巴组织细胞增多症(HLH)的死亡率很高。现有的风险评估方法存在不足,需要改进预测方法。本研究旨在使用 11 种不同的机器学习(ML)算法预测成人 HLH 患者的 30 天死亡率:对2015年1月至2021年9月期间的431名成人HLH患者进行了回顾性分析。使用最小绝对收缩和选择算子进行特征选择。我们采用了 11 种 ML 算法来创建预测模型。我们使用曲线下面积(AUC)、灵敏度、特异性、阳性预测值、阴性预测值、F1 评分、校准曲线和决策曲线分析来评估这些模型。我们使用 SHapley Additive exPlanation(SHAP)方法评估了特征的重要性:结果:七个独立预测因子成为最有价值的特征。在 11 种 ML 算法中,AUC 介于 0.65 和 1.00 之间。梯度提升决策树(GBDT)算法表现最佳(训练队列中为 1.00,验证队列中为 0.80)。通过使用 SHAP 方法,我们确定了对模型有贡献的变量及其与 30 天死亡率的相关性。当使用前 4 个特征(铁蛋白、UREA、年龄和凝血酶时间 (TT))时,GBDT 算法的 AUC 最高,在训练队列中达到 0.99,在验证队列中达到 0.83。此外,我们还开发了一个基于网络的计算器来估算30天的死亡风险:结论:将 GBDT 算法应用于实验室数据,可以准确预测 30 天死亡率。将这些算法融入临床实践可能会改善 30 天的预后。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Machine Learning of Laboratory Data in Predicting 30-Day Mortality for Adult Hemophagocytic Lymphohistiocytosis.

Background: Hemophagocytic Lymphohistiocytosis (HLH) carries a high mortality rate. Current existing risk-evaluation methodologies fall short and improved predictive methods are needed. This study aimed to forecast 30-day mortality in adult HLH patients using 11 distinct machine learning (ML) algorithms.

Methods: A retrospective analysis on 431 adult HLH patients from January 2015 to September 2021 was conducted. Feature selection was executed using the least absolute shrinkage and selection operator. We employed 11 ML algorithms to create prediction models. The area under the curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, F1 score, calibration curve and decision curve analysis were used to evaluate these models. We assessed feature importance using the SHapley Additive exPlanation (SHAP) approach.

Results: Seven independent predictors emerged as the most valuable features. An AUC between 0.65 and 1.00 was noted among the eleven ML algorithms. The gradient boosting decision tree (GBDT) algorithms demonstrated the most optimal performance (1.00 in the training cohort and 0.80 in the validation cohort). By employing the SHAP method, we identified the variables that contributed to the model and their correlation with 30-day mortality. The AUC of the GBDT algorithms was the highest when using the top 4 (ferritin, UREA, age and thrombin time (TT)) features, reaching 0.99 in the training cohort and 0.83 in the validation cohort. Additionally, we developed a web-based calculator to estimate the risk of 30-day mortality.

Conclusions: With GBDT algorithms applied to laboratory data, accurate prediction of 30-day mortality is achievable. Integrating these algorithms into clinical practice could potentially improve 30-day outcomes.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
12.20
自引率
9.90%
发文量
218
审稿时长
2 months
期刊介绍: The Journal of Clinical Immunology publishes impactful papers in the realm of human immunology, delving into the diagnosis, pathogenesis, prognosis, or treatment of human diseases. The journal places particular emphasis on primary immunodeficiencies and related diseases, encompassing inborn errors of immunity in a broad sense, their underlying genotypes, and diverse phenotypes. These phenotypes include infection, malignancy, allergy, auto-inflammation, and autoimmunity. We welcome a broad spectrum of studies in this domain, spanning genetic discovery, clinical description, immunologic assessment, diagnostic approaches, prognosis evaluation, and treatment interventions. Case reports are considered if they are genuinely original and accompanied by a concise review of the relevant medical literature, illustrating how the novel case study advances the field. The instructions to authors provide detailed guidance on the four categories of papers accepted by the journal.
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信