Two-centers machine learning analysis for predicting acid-fast bacilli results in tuberculosis sputum tests

IF 2 Q3 INFECTIOUS DISEASES

Journal of Clinical Tuberculosis and Other Mycobacterial Diseases Pub Date : 2025-02-01 DOI:10.1016/j.jctube.2025.100511

Jichong Zhu , Yong Zhao , Chengqian Huang , Chenxing Zhou , Shaofeng Wu , Tianyou Chen , Xinli Zhan

{"title":"Two-centers machine learning analysis for predicting acid-fast bacilli results in tuberculosis sputum tests","authors":"Jichong Zhu , Yong Zhao , Chengqian Huang , Chenxing Zhou , Shaofeng Wu , Tianyou Chen , Xinli Zhan","doi":"10.1016/j.jctube.2025.100511","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Tuberculosis (TB) is a chronic respiratory infectious disease caused by Mycobacterium tuberculosis, typically diagnosed through sputum smear microscopy for acid-fast bacilli (AFB) to assess the infectivity of TB.</div></div><div><h3>Methods</h3><div>This study enrolled 769 patients, including 641 patients from the First Affiliated Hospital of Guangxi Medical University as the training group, and 128 patients from Guangxi Hospital of the First Affiliated Hospital of Sun Yat-sen University as the validation group. Among the training cohort, 107 patients were AFB-positive, and 534 were AFB-negative. In the validation cohort, 24 were AFB-positive, and 104 were AFB-negative. Blood samples were collected and analyzed using machine learning (ML) methods to identify key factors for TB diagnosis.</div></div><div><h3>Results</h3><div>Several ML methods were compared, and support vector machine recursive feature elimination (SVM-RFE) was selected to construct a nomogram diagnostic model. The area under the curve (AUC) of the diagnostic model was 0.721 in the training cohort and 0.758 in the validation cohort. The model demonstrated clinical utility when the threshold was between 38% and 94%, with the NONE line above the ALL line in the decision curve analysis.</div></div><div><h3>Conclusion</h3><div>We developed a diagnostic model using multiple ML methods to predict AFB results, achieving satisfactory diagnostic performance.</div></div>","PeriodicalId":37942,"journal":{"name":"Journal of Clinical Tuberculosis and Other Mycobacterial Diseases","volume":"38 ","pages":"Article 100511"},"PeriodicalIF":2.0000,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Clinical Tuberculosis and Other Mycobacterial Diseases","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2405579425000026","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"INFECTIOUS DISEASES","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Tuberculosis (TB) is a chronic respiratory infectious disease caused by Mycobacterium tuberculosis, typically diagnosed through sputum smear microscopy for acid-fast bacilli (AFB) to assess the infectivity of TB.

Methods

This study enrolled 769 patients, including 641 patients from the First Affiliated Hospital of Guangxi Medical University as the training group, and 128 patients from Guangxi Hospital of the First Affiliated Hospital of Sun Yat-sen University as the validation group. Among the training cohort, 107 patients were AFB-positive, and 534 were AFB-negative. In the validation cohort, 24 were AFB-positive, and 104 were AFB-negative. Blood samples were collected and analyzed using machine learning (ML) methods to identify key factors for TB diagnosis.

Results

Several ML methods were compared, and support vector machine recursive feature elimination (SVM-RFE) was selected to construct a nomogram diagnostic model. The area under the curve (AUC) of the diagnostic model was 0.721 in the training cohort and 0.758 in the validation cohort. The model demonstrated clinical utility when the threshold was between 38% and 94%, with the NONE line above the ALL line in the decision curve analysis.

Conclusion

We developed a diagnostic model using multiple ML methods to predict AFB results, achieving satisfactory diagnostic performance.

查看原文本刊更多论文

双中心机器学习分析预测抗酸杆菌结核痰试验结果

结核病（TB）是一种由结核分枝杆菌引起的慢性呼吸道传染病，通常通过痰涂片镜检检测抗酸杆菌（AFB）来评估结核病的传染性。方法本研究共纳入769例患者，其中广西医科大学第一附属医院641例患者为试验组，中山大学第一附属广西医院128例患者为验证组。在培训队列中，107例患者为afb阳性，534例为afb阴性。在验证队列中，24例为afb阳性，104例为afb阴性。采集血液样本并使用机器学习（ML）方法进行分析，以确定结核病诊断的关键因素。结果对几种机器学习方法进行比较，选择支持向量机递归特征消除法（SVM-RFE）构建nomogram诊断模型。训练组和验证组诊断模型的曲线下面积（AUC）分别为0.721和0.758。当阈值在38% - 94%之间时，模型显示出临床实用性，决策曲线分析中NONE线高于ALL线。结论建立了一种多ML方法预测AFB结果的诊断模型，取得了满意的诊断效果。

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

Journal of Clinical Tuberculosis and Other Mycobacterial Diseases Medicine-Pulmonary and Respiratory Medicine

CiteScore

4.00

自引率

5.00%

发文量

审稿时长

30 weeks

期刊介绍： Journal of Clinical Tuberculosis and Mycobacterial Diseases aims to provide a forum for clinically relevant articles on all aspects of tuberculosis and other mycobacterial infections, including (but not limited to) epidemiology, clinical investigation, transmission, diagnosis, treatment, drug-resistance and public policy, and encourages the submission of clinical studies, thematic reviews and case reports. Journal of Clinical Tuberculosis and Mycobacterial Diseases is an Open Access publication.