Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with <italic style="font-style: italic">Talaromyces marneffei</italic> based on machine learning algorithms

Xie Yuling; Yang Kun; Dan Wenran

doi:10.13406/j.cnki.cyxb.003962

您当前的位置：

首页 >

文章列表页 >

Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with Talaromyces marneffei based on machine learning algorithms

Clinical research | 更新时间：2025-12-31

- Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with Talaromyces marneffei based on machine learning algorithms
- Journal of Chongqing Medical University Vol. 50, Issue 12, Pages: 1744-1750(2025)
- 作者机构：
  
  重庆市公共卫生医疗救治中心，重庆 400016
- 作者简介：
- 基金信息：
- DOI：10.13406/j.cnki.cyxb.003962
  CLC： R756
- Received：28 August 2024，
  
  Revised：2024-10-31，
  
  Accepted：23 December 2024，
  
  Published：28 December 2025
- 稿件说明：
移动端阅览
谢雨伶,杨坤,但文冉.基于机器学习算法构建HIV/AIDS合并马尔尼菲篮状菌感染诊断模型的探讨[J].重庆医科大学学报,2025,50(12):1744-1750.

Xie Yuling,Yang Kun,Dan Wenran.Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with Talaromyces marneffei based on machine learning algorithms[J].Journal of Chongqing Medical University,2025,50(12):1744-1750.
谢雨伶,杨坤,但文冉.基于机器学习算法构建HIV/AIDS合并马尔尼菲篮状菌感染诊断模型的探讨[J].重庆医科大学学报,2025,50(12):1744-1750. DOI： 10.13406/j.cnki.cyxb.003962.

Xie Yuling,Yang Kun,Dan Wenran.Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with Talaromyces marneffei based on machine learning algorithms[J].Journal of Chongqing Medical University,2025,50(12):1744-1750. DOI： 10.13406/j.cnki.cyxb.003962.

摘要

目的

基于机器学习算法构建人类免疫缺陷病毒/获得性免疫缺陷综合征（human immunodeficiency virus and acquired immunodeficiency syndrome，HIV/AIDS）患者合并马尔尼菲篮状菌（

Talaromyces marneffei，

TM）感染的诊断模型，以实现辅助早期诊断和提升诊断灵敏度。

方法

回顾性收集2020年1月至2023年9月在重庆市公共卫生医疗救治中心共201例HIV/AIDS-Mp1p抗原阳性患者实验室数据，筛选得到确诊TM（TM组）91例和未感染TM（非TM组）110例。将数据通过统计学分析获得2组间差异性指标，并构建logistic回归、随机森林分类器、决策树分类器模型。再通过最小绝对收缩和选择算子（least absolute shrinkage and selection operator，Lasso）回归筛选差异变量，并构建Lasso-logistic回归、随机森林和决策树分类器模型。分析所有模型筛选的高贡献特征指标，采用正确率、精确率、受试者工作特征（receiver operating characteristic，ROC）曲线和曲线下面积（area under the curve，AUC）评价模型的诊断性能。

结果

通过TM组与非TM组检验指标比较，删除了白细胞计数和性别2个差异无统计学意义的指标（

0.05），纳入红细胞计数、血小板计数、血红蛋白含量、C-反应蛋白、白细胞介素-6、降钙素原、年龄、CD4

T淋巴细胞计数、HIV-RNA含量、（1-3）-β-D葡聚糖检测、曲霉半乳甘露聚糖抗原检测11个差异有统计学意义（

0.05）的指标构建logistic回归模型、决策树和随机森林分类模型。进一步通过Lasso回归分析删除了CD4

T淋巴细胞计数和红细胞计数指标，剩余9个变量纳入构建了Lasso-logistic回归、决策树和随机森林分类模型，其AUC均大于纳入11个变量构建的同种模型，其中随机森林分类模型（

=9）的诊断性能最佳，正确率为0.797、精确率为0.794、AUC=0.822（95%CI=0.719~0.924）。

结论

在不同诊断模型中实验室检验指标特征重要性不同，经Lasso回归筛选变量后再构建模型能提高诊断性能，构建的所有模型中随机森林分类模型诊断性能最好。基于机器学习算法和临床检验数据建立诊断模型，有利于辅助临床早期诊断HIV/AIDS合并TM感染。

Abstract

Objective

To establish

a diagnostic model for human immunodeficiency virus/acquired immunodeficiency syndrome（HIV/AIDS） patients co-infected with

Talaromyces marneffei

（TM） using machine learning algorithms，and to help realize early diagnosis and improve diagnostic sensitivity.

Methods

A retrospective analysis was performed for the laboratory test data of 201 HIV/AIDS patients with positive Mp1p antigen from January 2020 to September 2023，among whom 91 were diagnosed with TM infection（TM group） and 110 were not infected with TM（non-TM group）. The statistical analysis was performed to identify differential indicators between the two groups，and the models of logistic regression，Random Forest Classifier，and Decision Tree Classifier were constructed. The Least absolute shrinkage and selection operator（LASSO） regression was used to identify differential variables，and then the models of LASSO-logistic regression，Random Forest Classifier，and Decision Tree Classifier were constructed. The characteristic indicators with high contribution identified by all models were analyzed，and the diagnostic performance of the model was assessed by accuracy，precision，the receiver operating characteristic（ROC） curve，and the area under the ROC curve（AUC）.

Results

Through comparison of test indicators between the TM group and the non-TM group，two indicators without statistical significance（leukocyte count and sex） were excluded（

0.05），and eleven indicators with statistical significance were used to establish the models of logistic regression，Decision Tree Classifier，and Random Forest Classifier，i.e.，red blood cell count，platelet count，hemoglobin content，C-reactive protein，interleukin-6，procalcitonin，age，CD4

T lymphocyte count，HIV-RNA level，（1-3）-β-D-glucan test，and

Aspergillus galactomannan

antigen test（

0.05）. The indicators of CD4

T lymphocyte count and red blood cell count were removed by the LASSO r

egression analysis，and the remaining nine variables were included to establish the models of LASSO-logistic regression，Decision Tree Classifier，and Random Forest Classifier，which had a greater AUC than the models established based on the eleven variables； the Random Forest Classifier model（

=9） had the best diagnostic performance，with an accuracy rate of 0.797，a precision rate of 0.794，and an AUC of 0.822（95%CI=0.719-0.924）.

Conclusion

Laboratory test indicators have different levels of importance in different diagnostic models，and the model established after screening for variables based on LASSO regression can improve the diagnostic performance of the model. The Random Forest Classifier model has the best diagnostic performance among all the models established. The establishment of diagnostic model based on machine learning algorithms and clinical test data can help with the early diagnosis of HIV/AIDS with TM infection in clinical practice.

关键词

Keywords

references

Cen JM ， Huang J ， Zeng W ， et al . Clinical features of HIV positive talaromycosis marneffei patients and development of a risk prediction model [J ] . Heliyon ， 2023 ， 9 （ 9 ）： e20068 ．

Zhang ZH ， Li BP ， Chai ZL ， et al . Evolution of the ability to evade host innate immune defense by Talaromyces marneffei [J ] . Int J Biol Macromol ， 2023 ， 253 （ Pt 8 ）： 127597 ．

Son VT ， Khue PM ， Strobel M . Penicilliosis and AIDS in Haiphong，Vietnam：evolution and predictive factors of death [J ] . Med Mal Infect ， 2014 ， 44 （ 11/12 ）： 495 - 501 ．

Wang F ， Han RH ， Chen S . An overlooked and underrated endemic mycosis-talaromycosis and the pathogenic fungus Talaromyces marneffei [J ] . Clin Microbiol Rev ， 2023 ， 36 （ 1 ）： e0005122 ．

李玥琪 . 机器学习模型在HIV合并马尔尼菲篮状菌早期诊断中的应用研究 [D ] . 南宁：广西医科大学， 2021 ．

Li YQ . Application of machine learning model in early diagnosis of HIV complicated with basilaria marneffei [D ] . Nanning ： Guangxi Medical University ， 2021 ．

Peng L ， Shi Y ， Zheng L ， et al . Clinical features of patients with talaromycosis marneffei and microbiological characteristics of the causative strains [J ] . Journal of Clinical Laboratory Analysis ， 2022 ， 36 （ 11 ）： e24737 .

Chen XM ， Ou X ， Wang HD ， et al . Talaromyces marneffei Mp1p antigen detection may play an important role in the early diagnosis of talaromycosis in patients with acquired immunodeficiency syndrome [J ] . Mycopathologia ， 2022 ， 187 （ 2/3 ）： 205 - 215 ．

Wei HY ， Liang WJ ， Li B ， et al . Clinical characteristics and risk factors of Talaromyces marneffei infection in human immunodeficiency virus-negative patients：a retrospective observational study [J ] . World J Emerg Med ， 2021 ， 12 （ 4 ）： 281 - 286 ．

Jiang J ， Meng S ， Huang S ， et al . Effects of Talaromyces marneffei infection on mortality of HIV/AIDS patients in Southern China：a retrospective cohort study [J ] . Clin Microbiol Infect ， 2019 ， 25 （ 2 ）： 233 - 241 ．

杜　然，滕　腾，赵云凤，等 . Lasso-Logistic回归模型拟合临床因素、NF-κB/NLRP3信号通路预测心肌梗死后缺血性心肌病价值 [J ] . 中国急救复苏与灾害医学杂志， 2024 ， 19 （ 6 ）： 1015 - 1023 ．

Du R ， Teng T ， Zhao YF ， et al . Lasso-Logistic regression model fitting clinical factors and NF-κB/NLRP3 predictive value of ischemic cardiomyopathy after myocardial infarction [J ] . China J Emerg Resusc Disaster Med ， 2024 ， 19 （ 6 ）： 1015 - 1023 ．

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research advances in artificial intelligence for difficult airway prediction

Risk prediction of early esophageal varices in patients with liver cirrhosis based on interpretable machine learning

Optimization strategy and practice analysis of airway management based on artificial intelligence

Construction of an online interactive calculation tool and corresponding risk stratification system for the probability of postoperative anastomotic leak in esophageal atresia based on machine learning algorithms

Related Author

Xie Yuling

Shao Jianlin

Zhou Guilin

Yin Yuheng

Yang Chun

Fan Jie

Wang Yuwen

Wang Wei

Related Institution

Department of Anesthesiology，The First Affiliated Hospital of Kunming Medical University

Department of Infection Diseases， The First Affiliated Hospital of Chongqing Medical University

College of Public Health，Chongqing Medical University

School of Public Health Emergency Management，Chongqing Medical and Pharmaceutical College

Department of Anesthesiology， Peking University Third Hospital

⁰

Establishment of a diagnostic model for human immunodeficiency virus/ acquired immunodeficiency syndrome co-infection with Talaromyces marneffei based on machine learning algorithms

DOI：10.13406/j.cnki.cyxb.003962

摘要

Abstract

关键词

Keywords

references