基于机器学习的慢阻肺患者再入院预测研究

doi:10.12052/gdutxb.230206

摘要/Abstract

摘要： 由于慢性阻塞性肺病(简称慢阻肺)的高复发性，患者计划外再入院问题已成为严峻挑战。本文提出融合不同结构化数据和多种机器学习算法进行风险预测的框架和方法，并以广州某三甲医院近万名慢阻肺患者的真实电子病历数据进行演示。通过构建双向长短期记忆条件随机场命名实体识别模型处理非结构化信息，使用支持向量机、随机森林、极限梯度提升机和反向传播神经网络构建风险预测模型，发现极限梯度提升机模型的预测性能最佳，以及住院时长、查尔森合并症指数、病程、白细胞和嗜酸性粒细胞是再入院最重要的影响因素。本文研究丰富了慢阻肺的相关知识，并为其早期发现、及时诊断和精准干预提供了研究思路和辅助决策工具。

关键词: 慢阻肺, 再入院预测, 命名实体识别, 机器学习

Abstract: Due to the high recurrence rate of chronic obstructive pulmonary disease (COPD), the issue of unplanned readmissions has become a significant challenge for patients. In this research, a framework and a methodology are proposed that integrate different structured data and multiple machine learning algorithms for risk prediction. A method is showed using genuine electronic medical information from approximately 10,000 COPD patients at a tertiary hospital in Guangzhou, China. To handle unstructured input, a Bidirectional Long Short-Term Memory-Conditional Random Field (BiLSTM-CRF) model known as named entity recognition is used. Furthermore, risk prediction models with Support Vector Machines (SVM), Random Forests (RF), Extreme Gradient Boosting (XGBoost), and Back Propagation (BP) Neural Network are developed. The results show that the XGBoost model performs best. The length of hospital stay, Charlson Comorbidity Index, disease duration, white blood cell count, and eosinophil count are also identified as the most relevant predictors for readmission. An understanding of chronic diseases is advanced by providing research insights and decision support tools for early detection, prompt diagnosis, and precise intervention.

Key words: chronic obstructive pulmonary disease (COPD), readmission prediction, named entity recognition, machine learning

中图分类号:

TP393

吴菊华, 郑稳, 聂亚, 陶雷. 基于机器学习的慢阻肺患者再入院预测研究[J]. 广东工业大学学报,doi: 10.12052/gdutxb.230206

Wu Ju-hua, Zheng Wen, Nie Ya, Tao Lei. Readmission Prediction for Patients with Chronic Obstructive Pulmonary Disease Based on Machine Learning[J]. Journal of Guangdong University of Technology, 2024, 41(0): 0-.doi: 10.12052/gdutxb.230206

参考文献

[1] RUAN H, ZHANG H, WANG J, et al. Readmission rate for acute exacerbation of chronic obstructive pulmonary disease: a systematic review and meta-analysis [J]. Respiratory Medicine, 2023, 206(3): 1070-1082.
[2] 窦一峰, 吕劲松, 白景珍, 等. 1659例慢性阻塞性肺疾病出院患者的统计分析[J]. 中国病案, 2023, 24(2): 80-83.
DOU Y F, LYU J S, BAI J Z, et al. Statistical analysis of 1659 discharged patient of chronic obstructive pulmonary disease [J]. Chinese Medical Record, 2023, 24(2): 80-83.
[3] JIANG S, CHIN K, QU G, et al. An integrated machine learning framework for hospital readmission prediction [J]. Knowledge-Based Systems, 2018, 146(15): 73-90.
[4] HAITL S, LOPEZ-CAMPOS J L, POZO-RODRIGUEZ F, et al. Risk of death and readmission of hospital-admitted COPD exacerbations: European COPD Audit [J]. European Respiratory Journal, 2016, 47(1): 113-121.
[5] 蔡柏蔷, 陈荣昌. 慢性阻塞性肺疾病急性加重(AECOPD) 诊治中国专家共识(2017年更新版) [J]. 国际呼吸杂志, 2017, 37(14): 1041-1057.
CAI B Q, CHEN R C. Chinese expert consensus on diagnosis and treatment of acute exacerbation of chronic obstructive pulmonary disease (AECOPD) [J]. International Journal of Respiration, 2017, 37(14): 1041-1057.
[6] PRESS V G, AU D H, BOURBEAU J, et al. Reducing chronic obstructive pulmonary disease hospital readmissions: an official American Thoracic Society workshop report [J]. Annals of the American Thoracic Society, 2019, 16(2): 161-170.
[7] CHOI E, BAHADORI M T, SCHUETZ A, et al. Doctor ai: predicting clinical events via recurrent neural networks[C]//Machine Learning for Healthcare Conference. New York: PMLR, 2016: 301-318.
[8] CHOI E, BAHADORI M T, SUN J, et al. Retain: An interpretable predictive model for healthcare using reverse time attention mechanism [J]. Advances in Neural Information Processing Systems, 2016, 29(7): 3512-3520.
[9] 崔少泽, 赵森尧, 王延章. 基于ADASYN-IFA-Stacking的再入院患者风险预测方法[J]. 系统工程理论与实践, 2021, 41(3): 744-758.
CUI S Z, ZHAO S Y, WANG Y Z. Risk prediction method for readmission patients based on ADASYN-IFA-Stacking [J]. Systems Engineering–Theory & Practice, 2021, 41(3): 744-758.
[10] LI J, LIANG L, CAO S, et al. Secular trend and risk factors of 30-day COPD-related readmission in Beijing, China [J]. Scientific Reports, 2022, 12(1): 165-174.
[11] LEONG K T G, WONG L Y, AUNG K C Y, et al. Risk stratification model for 30-day heart failure readmission in a multiethnic South East Asian community [J]. The American Journal of Cardiology, 2017, 119(9): 1428-1432.
[12] HSU J C, WU F H, LIN H H, et al. AI models for predicting readmission of pneumonia patients within 30 days after discharge [J]. Electronics, 2022, 11(5): 673-681.
[13] 黄光成, 周良, 石建伟, 等. 机器学习算法在疾病风险预测中的应用与比较[J]. 中国卫生资源, 2020, 23(4): 432-436.
HUANG G C, ZHOU L, SHI J W, et al. Application and comparison of machine learning algorithms in disease risk prediction [J]. Chinese Health Resources, 2020, 23(4): 432-436.
[14] KRISHNAMOORTHI R, JOSHI S, ALMARZOUKI H Z, et al. A novel diabetes healthcare disease prediction framework using machine learning techniques [J]. Journal of Healthcare Engineering, 2022(3): 1-10.
[15] LI M, CHENG K, KU K, et al. Modelling 30-day hospital readmission after discharge for COPD patients based on electronic health records [J]. NPJ Primary Care Respiratory Medicine, 2023, 33(1): 16-28.
[16] LI J, MA X, ZENG X, et al. Risk factors of readmission within 90 days for chronic obstructive pulmonary disease patients with frailty and construction of an early warning model [J]. International Journal of Chronic Obstructive Pulmonary Disease, 2023(18): 975-984.
[17] BONOMO M, HERMSEN M G, KASKOVICH S, et al. Using machine learning to predict likelihood and cause of readmission after hospitalization for chronic obstructive pulmonary disease exacerbation [J]. International Journal of Chronic Obstructive Pulmonary Disease, 2022(17): 2701-2709.
[18] ZHANG R, LU H, CHANG Y, et al. Prediction of 30-day risk of acute exacerbation of readmission in elderly patients with COPD based on support vector machine model [J]. BMC Pulmonary Medicine, 2022, 22(1): 292-305.
[19] JO Y S, RHEE C K, KIM K J, et al. Risk factors for early readmission after acute exacerbation of chronic obstructive pulmonary disease [J]. Therapeutic Advances in Respiratory Disease, 2020, 14(3): 1-11.
[20] 张瑞, 吴珍珍, 常艳, 等. 老年慢性阻塞性肺疾病患者 30 天内急性加重再入院风险预测模型的构建与验证[J]. 中国呼吸与危重监护杂志, 2021, 20(7): 457-464.
ZHANG R, WU Z Z, CHANG Y, et al. Construction and validation of a predictive model of acute exacerbation readmission risk within 30 days in elderly patients with chronic obstructive pulmonary disease [J]. Chinese Journal of Respiratory and Critical Care Medicine, 2021, 20(7): 457-464.
[21] GOTO T, JO T, MATSUI H, et al. Machine learning-based prediction models for 30-day readmission after hospitalization for chronic obstructive pulmonary disease [J]. COPD: Journal of Chronic Obstructive Pulmonary Disease, 2019, 16(5-6): 338-343.
[22] BOSE P, SRINIVASAN S, SLEEMAN W C, et al. A survey on recent named entity recognition and relationship extraction techniques on clinical texts [J]. Applied Sciences, 2021, 11(18): 8319-8321.
[23] HEGEWALD M J, HORNE B D, TRUDO F, et al. Blood eosinophil count and hospital readmission in patients with acute exacerbation of chronic obstructive pulmonary disease [J]. International Journal of Chronic Obstructive Pulmonary Disease, 2020, 15(13): 2629-2641.
[24] PETRAZZINI B O, CHAUDHARY K, MARQUEZ-LUNA C, et al. Coronary risk estimation based on clinical data in electronic health records [J]. Journal of the American College of Cardiology, 2022, 79(12): 1155-1166.
[25] LI B, ZHANG Y, WU X. DLKN-MLC: a disease prediction model via multi-label learning [J]. International Journal of Environmental Research and Public Health, 2022, 19(15): 9771-9780.
[26] SUNDARARAMAN A, RAMANATHAN S V, THATI R. Novel approach to predict hospital readmissions using feature selection from unstructured data with class imbalance [J]. Big Data Research, 2018, 13(7): 65-75.
[27] SALEM H, RUIZ A, HERNANDEZ S, et al. Borderline personality features in inpatients with bipolar disorder: impact on course and machine learning model use to predict rapid readmission [J]. Journal of Psychiatric Practice, 2019, 25(4): 279-289.
[28] MOREL D, KALVIN C Y, LIU-FERRARA A, et al. Predicting hospital readmission in patients with mental or substance use disorders: a machine learning approach [J]. International Journal of Medical Informatics, 2020, 139(15): 104-136.
[29] FENG H Q, SUN Y, YANG S, et al. Chinese electronic medical record named entity recognition based on BERT methods [J]. Computer Engineering and Design, 2023, 44(4): 1220-1227.
[30] ElDIN H G, ABDULRAZEK M, ABDELSHAFI M, et al. Med-Flair: medical named entity recognition for diseases and medications based on Flair embedding [J]. Procedia Computer Science, 2021, 189(5): 67-75.
[31] KE J, WANG W, CHEN X, et al. Medical entity recognition and knowledge map relationship analysis of Chinese EMRs based on improved BiLSTM-CRF [J]. Computers and Electrical Engineering, 2023, 108(3): 1087-1092.
[32] DAI Z, WANG X, NI P, et al. Named entity recognition using BERT BiLSTM CRF for Chinese electronic health records[C]//2019 12th international congress on image and signal processing, biomedical engineering and informatics. Suzhou: CISP-BMEI, 2019: 1-5.
[33] CUI Z, YUAN Z, WU Y, et al. Intelligent recommendation for departments based on medical knowledge graph [J]. IEEE Access, 2023, 11(4): 25372-25385.
[34] LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions [J]. Advances in Neural Information Processing Systems, 2017(30): 4768-4777.
[35] CHENG S L, LI Y R, HUANG N, et al. Effectiveness of nationwide COPD pay-for-performance program on COPD exacerbations in Taiwan [J]. International Journal of Chronic Obstructive Pulmonary Disease, 2021(16): 2869-2881.
[36] JABARKHIL A, MOBERG M, JANNER J, et al. Elevated blood eosinophils in acute COPD exacerbations: better short-and long-term prognosis [J]. European Clinical Respiratory Journal, 2020, 7(1): 175-183.
[37] ALQAHTANI J S, ALDABAYAN Y S, ALDHAHIR A M, et al. Predictors of 30-and 90-day COPD exacerbation readmission: a prospective cohort study [J]. International Journal of Chronic Obstructive Pulmonary Disease, 2021, 16(5): 2769-2781.
[38] BERNABEU-MORA R, GARCIA-GUILLAMON G, VALERA-NOVELLA E, et al. Frailty is a predictive factor of readmission within 90 days of hospitalization for acute exacerbations of chronic obstructive pulmonary disease: a longitudinal study [J]. Therapeutic Advances in Respiratory Disease, 2017, 11(10): 383-392.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed