基于机器学习的慢阻肺急性加重疾病预测与疾病标志物筛选研究

于淼; 郭金兴; 孙悦; 陈广新; 周志尊; 孙延斌

doi:10.3969/j.issn.2096-6091.2022.24.003

您当前的位置：

首页 >

文章列表页 >

基于机器学习的慢阻肺急性加重疾病预测与疾病标志物筛选研究

研究论文 | 更新时间：2023-08-03

- 基于机器学习的慢阻肺急性加重疾病预测与疾病标志物筛选研究
- Study on Prediction of Acute Exacerbation of COPD and Screening of Disease Markers Based on Machine Learning
- 新一代信息技术 2022年5卷第24期页码：12-17
- 作者机构：
  
  1.牡丹江医学院，黑龙江牡丹江 157011
  2.牡丹江医学院附属红旗医院，黑龙江牡丹江 157011
- 作者简介：
  
  于淼（1997‒），女，硕士研究生，研究方向：医学人工智能。
  陈广新（1978‒），男，硕士，讲师，研究方向：机器学习。
  孙延斌（1974‒），男，硕士，副教授，研究方向：临床医学。E-mail： vvvcgx@126.com
- 基金信息：
  
  黑龙江省高等学校基本科研业务费项目(2020-KYYWFMY-0052);牡丹江医学院研究生导师科研专项计划项目(YJSZX2022135)
- DOI：10.3969/j.issn.2096-6091.2022.24.003
  中图分类号：
扫描看全文
于淼, 郭金兴, 孙悦, 等. 基于机器学习的慢阻肺急性加重疾病预测与疾病标志物筛选研究[J]. 新一代信息技术, 2022, 5(24): 12-17

YU Miao, GUO Jin-xing, SUN Yue, et al. Study on Prediction of Acute Exacerbation of COPD and Screening of Disease Markers Based on Machine Learning[J]. New Generation of Information Technology, 2022, 5(24): 12-17
于淼, 郭金兴, 孙悦, 等. 基于机器学习的慢阻肺急性加重疾病预测与疾病标志物筛选研究[J]. 新一代信息技术, 2022, 5(24): 12-17 DOI： 10.3969/j.issn.2096-6091.2022.24.003.

YU Miao, GUO Jin-xing, SUN Yue, et al. Study on Prediction of Acute Exacerbation of COPD and Screening of Disease Markers Based on Machine Learning[J]. New Generation of Information Technology, 2022, 5(24): 12-17 DOI： 10.3969/j.issn.2096-6091.2022.24.003.

摘要

针对慢性阻塞性疾病急性加重患者临床诊断的困难、准确性差的问题，提出基于机器学习构建临床预测模型和SHAP模型解释性筛选疾病关键特征，通过不同机器学习模型的预测性能比较与特征重要性、模型解释性分析，找到最优的模型与疾病标志特征。选取伦敦大学学院患者数据库COPD数据集99例。利用6种机器学习算法建立预测模型，采用k折交叉验证及ROC-AUC、F1分数等评价指标，对模型进行验证。应用SHAP对预测模型进行解释性分析。综合各个模型的表现，SVC模型相较于其他模型预测性能最佳，准确率达到0.93，AUC为0.99，FEV1PRED和FEV1是影响COPDSEVERITY的重要因素。基于机器学习预测模型可以对COPDSEVERITY诊断提供有效的辅助决策支持。

Abstract

In response to the difficulties and poor accuracy of clinical diagnosis for patients with acute exacerbation of chronic obstructive pulmonary disease, we propose to construct a clinical prediction model based on machine learning and use SHAP model interpretability to screen key disease features. By comparing the predictive performance and feature importance of different machine learning models and conducting model interpretability analysis, we aim to find the optimal model and disease marker features. We selected a dataset of 99 patients with COPD from the University College London patient database. We used six machine learning algorithms to establish a prediction model and evaluated the model using k-fold cross-validation and evaluation metrics such as ROC-AUC and f1 score. We then used SHAP for interpretability analysis of the prediction model. Among all the models, the SVC model had the best predictive performance, with an accuracy of 0.93 and AUC of 0.99. FEV1PRED and FEV1 were identified as important factors affecting COPDSEVERITY. The machine learning prediction model can provide effective decision-making support for the diagnosis of COPDSEVERITY.

关键词

机器学习慢阻肺预测模型模型解释

Keywords

lower respiratory tract modelingchronic obstructive pulmonary diseasecomputational fluid dynamicsnumerical simulation

references

FERNANDEZ-GRANERO M A, SANCHEZ-MORILLOD D, LEON-JIMENEZ A. An artificial intelligence approach to early predict symptom-based exacerbations of COPD[J]. Biotechnology & Biotechnological Equipment, 2018, 32(3): 778-784.

MOHKTAR M S, REDMOND S J, ANTONIADES N C, et al. Predicting the risk of exacerbation in patients with chronic obstructive pulmonary disease using home telehealth measurement data[J]. Artificial Intelligence in Medicine, 2015, 63(1): 51-59.

SHAH S A, VELARDO C, FARMER A, et al. Exacerbations in chronic obstructive pulmonary disease: Identification and prediction using a digital health system[J]. Journal of Medical Internet Research, 2017, 19(3): e69.

VERMA V K, LIN W Y. A machine learning-based predictive model for 30-day hospital readmission prediction for COPD patients[C]//2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC). Piscataway: IEEE, 2020: 994-999.

PAN W H, CHEN J Y, HAUNG S L, et al. Reference spirometric values in healthy Chinese neversmokers in two townships of Taiwan, China[J]. The Chinese Journal of Physiology, 1997, 40(3): 165-174.

IP M S, WAI-SAN KO F, LAU A C, et al. Updated spirometric reference values for adult Chinese in Hong Kong, China and implications on clinical utilization[J]. Chest, 2006, 129(2): 384-392.

DUONG M, ISLAM S, RANGARAJAN S, et al. Global differences in lung function by region (PURE): An International, community-based prospective study[J]. Lancet Respiratory Medicine, 2013, 1(8): 599-609.

MCKEON F, XIAN W, RAO W, et al. Regenerative metaplastic clones in COPD lung drive inflammation and fibrosis[J]. Cell, 2020, 181(4): 848-864.

JIA T G, ZHAO J Q, LIU J H. Serum inflammatory factor and cytokines in AECOPD [J]. Asian Pacific Journal of Tropical Medicine, 2014, 7(12): 1005-1008.

KARADENIZ G, POLAT G, SENOL G, et al. C-reactive protein measurements as a marker of the severity of chronic obstructive pulmonary disease exacerbations[J]. Inflammation, 2013, 36(4): 948-953.

PIKOULA M, QUINT J K, NISSEN F, et al. Identifying clinically important COPD sub-types using data-driven approaches in primary care population based electronic health records[J]. Medical Informatics and Decision Making, 2019, 19(1): 86.

WANG C, CHEN X, DU L, et al. Comparison of machine learning algorithms for the identification of acute exacerbations in chronic obstructive pulmonary disease[J]. Computer Methods and Programs in Biomedicine, 2020, 188: 105267.

CHEN J, YANG Z, YUAN Q, et al. Prediction models for pulmonary function during acute exacerbation of chronic obstructive pulmonary disease[J]. Physiological Measurement, 2021, 41(12): 125010.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

预测COVID-19感染风险的医疗决策机器学习方法研究

基于改进支持向量回归算法的甲状腺疾病预测

基于可解释性机器学习的新冠肺炎疾病风险预测研究