可解释的机器学习用于预测艾滋病毒/结核病联合感染的临床结果:一项比较后期研究
Qingfeng Sun1, Kai Zhang2, Yuanlong Xu2
1Department of Tuberculosis, Guangxi Zhuang Autonomous Region Chest Hospital, No 8, Yangjiaoshan Road, Liuzhou, Guangxi, 545005, P. R. China.
机器学习准确地预测了艾滋病毒和结核病联合感染的患者的结果. 该工具有助于识别高风险个体,以提高治疗成功率.
科学领域:
- 医疗信息学 医疗信息学
- 公共卫生 公共卫生
- 机器学习在医学中的应用
背景情况:
- 与单独结核病相比,艾滋病毒/结核病联合感染显著增加了死亡率和治疗失败率.
- 机器学习 (ML) 为早期识别高风险患者提供了先进的方法.
- 早期风险分层对于改善同时感染人群中的患者结果至关重要.
研究的目的:
- 开发和验证一种机器学习模型,用于预测艾滋病毒/结核病共感染患者的结果.
- 确定不良结果的关键临床和免疫学预测因素.
- 评估ML工具在临床实践中的资源配置和治疗优化的潜力.
主要方法:
- 对359名同时感染艾滋病毒/结核病的患者的回顾性分析.
- 使用了六个ML分类器 (随机森林,XGBoost,LightGBM,SVM,额外树木,CatBoost) 与SMOTE进行类不平衡.
- 模型性能使用AUC,准确性,精度,回忆,F1得分进行评估,并使用TOPSIS进行排名;领先模型使用SHAP进行解释.
主要成果:
- 轻GBM模型实现了最高的性能 (AUC=0.771,精度=84.72%,F1=0.522).
- SHAP分析确定了年龄,CD4/CD8计数,BMI和职业作为重要的预测因素.
- 较低的BMI,严重的免疫抑制和年龄较大与不利的结果有关.
结论:
- 一个带有SHAP解释的梯度增强模型 (LightGBM) 可靠地预测了HIV/结核病共感染的结果.
- 该模型突出了临床可操作的风险因素,使得高风险患者的早期识别成为可能.
- 将其整合到临床工作流程中可以提高资源配置,并提高结核病治疗的成功率.
更多相关视频
07:06Prediction of HIV-1 Coreceptor Usage Tropism by Sequence Analysis using a Genotypic Approach
Published on: December 1, 2011
06:22Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
Published on: September 19, 2025
相关概念视频
Pulmonary Tuberculosis II
Here is a detailed explanation of its pathophysiology:
Transmission: The process begins when a person inhales droplet nuclei containing M. tuberculosis. These are typically released into the air when an individual with pulmonary or...
Pulmonary Tuberculosis I
Causative Organism
The primary infectious agent causing tuberculosis is Mycobacterium tuberculosis, a slow-growing, acid-fast, aerobic rod that exhibits sensitivity to heat and ultraviolet light. Instances of Mycobacterium bovis and Mycobacterium avium contributing to the development of TB infection are rare.
Mode of...
Statistical Methods for Analyzing Epidemiological Data
Steps in Outbreak Investigation
Bias in Epidemiological Studies
Pulmonary Tuberculosis III
The first classification is based on the development of the disease, and it includes the following categories:
