您的位置：山东大学 -> 科技期刊社 -> 《山东大学学报（医学版）》

山东大学学报 (医学版) ›› 2026, Vol. 64 ›› Issue (3): 108-115.doi: 10.6040/j.issn.1671-7554.0.2024.0940

• 临床医学 • 上一篇    

基于动态列线图及机器学习的慢性阻塞性肺疾病急性加重期伴发肺性脑病风险预测模型构建及验证

陆晨琳,许露,杨俊发,潘华琴,倪清涛   

  • 发布日期:2026-03-19
  • 通讯作者: 许露. E-mail:xl1807501296@126.com
  • 基金资助:
    江苏省双创博士项目(202030205)

Construction and validation of a risk prediction model for pulmonary encephalopathy associated with acute exacerbation of chronic obstructive pulmonary disease based on dynamic nomogram and machine learning

LU Chenlin, XU Lu, YANG Junfa, PAN Huaqin, NI Qingtao   

  1. Department of Respiratory and Critical Care Medicine, Taizhou Peoples Hospital, Taizhou 225300, Jiangsu, China
  • Published:2026-03-19

PDF (PC)

3

摘要： 目的 基于动态列线图及机器学习构建预测慢性阻塞性肺疾病急性加重期(acute exacerbation of chronic obstructive pulmonary disease, AECOPD)伴发肺性脑病风险的模型,并对其预测效能进行验证。 方法 选取2022年1月至2024年6月泰州市人民医院收治的272例AECOPD患者作为研究对象,根据是否伴发肺性脑病将患者分为肺性脑病组(n=54)和非肺性脑病组(n=218),使用单因素和多因素logistic回归分析AECOPD伴发肺性脑病的危险因素,并建立相关预测模型。 结果 单因素分析结果显示,两组患者性别、年龄、体质量指数(body mass index, BMI)、吸烟史、高血压、糖尿病、高血脂、慢性阻塞性肺疾病(chronic obstructive pulmonary disease, COPD)病程、急性加重次数、心率和血清钠比较,差异无统计学意义(P>0.05);两组血氧分压(partial pressure of oxygen, PaO2)、血二氧化碳分压(partial pressure of carbon dioxide, PaCO2)、血pH、血清钾、血清白蛋白和C-反应蛋白(C-reactive protein, CRP)比较,差异有统计学意义(P<0.05)。多因素分析结果显示,PaO2较低、PaCO2较高、血pH值较低、血清钾较低和CRP较高是AECOPD患者伴发肺性脑病的独立危险因素(P<0.05)。动态列线图的Hosmer-Lemeshow拟合度检验结果:χ2 =2.912, P=0.940,随机森林的Hosmer-Lemeshow拟合度检验结果:χ2 =12.628, P=0.125,决策树模型的Hosmer-Lemeshow拟合度检验结果:χ2 =9.232, P=0.323;ROC曲线的AUC分别为0.874(95%CI:0.822~0.925)、0.802(95%CI:0.727~0.877)和0.847(95%CI:0.788~0.905)。 结论 基于危险因素构建的动态列线图、随机森林和决策树模型均能够有效预测AECOPD患者伴发肺性脑病的风险。

关键词: 慢性阻塞性肺疾病急性加重期, 肺性脑病, 动态列线图, 随机森林, 决策树

Abstract: Objective To construct a model for predicting the risk of acute exacerbation of chronic obstructive pulmonary disease(AECOPD)associated with pulmonary encephalopathy based on dynamic nomogram and machine learning, and to verify the predictive efficacy of the model. Methods A total of 272 patients with AECOPD admitted to Taizhou Peoples Hospital from January 2022 to June 2024 were selected as research objects. According to whether they were accompanied by pulmonary encephalopathy, the patients were divided into pulmonary encephalopathy group(n=54)and non-pulmonary encephalopathy group(n=218). Univariate and multivariate logistic regression were used to analyze the risk factors of AECOPD associated with pulmonary encephalopathy, and the related prediction models were established. Results Univariate analysis showed that there was no significant difference in gender, age, body mass index(BMI), smoking history, hypertension, diabetes, hyperlipidemia, chronic obstructive pulmonary disease(COPD)course, acute exacerbations, heart rate and serum sodium between the two groups(P>0.05). There were significant differences in partial pressure of oxygen(PaO2), partial pressure of carbon dioxide(PaCO2), blood pH, serum potassium, serum albumin and C-reactive protein(CRP)between the two groups(P<0.05). The results of multifactor analysis showed that lower PaO2, higher PaCO2, lower blood pH, lower serum potassium and higher CRP levels were independent risk factors for pulmonary encephalopathy in AECOPD patients(P<0.05). Hosmer-Lemeshow fit test showed the following results: for dynamic nomogram, χ2=2.912, P=0.940; for random forest, χ2=12.628, P=0.125; for decision tree models, χ2=9.232, P=0.323. The AUCs were 0.874(95%CI: 0.822-0.925), 0.802(95%CI: 0.727-0.877)and 0.847(95%CI: 0.788-0.905), respectively. Conclusion Dynamic nomogram, random forest and decision tree models based on risk factors can effectively predict the risk of pulmonary encephalopathy in patients with AECOPD.

Key words: Acute exacerbation of chronic obstructive pulmonary disease, Pulmonary encephalopathy, Dynamic nomogram, Random forest, Decision tree

中图分类号: 

  • R563
[1] 王凤燕, 梁振宇, 李雪萍, 等. 慢性阻塞性肺疾病近年临床研究热点[J]. 山东大学学报(医学版), 2024, 62(5): 7-15. WANG Fengyan, LIANG Zhenyu, LI Xueping, et al. Clinical research hotspots of chronic obstructive pulmonary disease in recent years [J]. Journal of Shandong University(Health Sciences), 2024, 62(5): 7-15.
[2] Pratt AJ, Purssell A, Zhang T, et al. Complexity in clinical diagnoses of acute exacerbation of chronic obstructive pulmonary disease[J]. BMC Pulm Med, 2023, 23(1): 298. doi: 10.1186/s12890-023-02587-1
[3] 林果为,王言耀,葛均波.实用内科学[M].北京:人民卫生出版社, 2017: 1204.
[4] 慢性阻塞性肺疾病急性加重诊治专家组. 慢性阻塞性肺疾病急性加重诊治中国专家共识(2023年修订版)[J]. 国际呼吸杂志, 2023, 43(2): 132-149. Expert Group on the Diagnosis and Treatment of Acute Exacerbation of Chronic Obstructive Pulmonary Disease. Expert consensus on the acute exacerbation of chronic obstructive pulmonary disease in China(revision in 2023)[J]. International Journal of Respiratory Diseases, 2023, 43(2): 132-149.
[5] 李占东, 佟杰. 中性粒细胞/淋巴细胞比值对AECOPD发展为肺性脑病风险的预测价值[J]. 大连医科大学学报, 2022, 44(6): 511-514. LI Zhandong, TONG Jie. Predictive value of neutrophil/lymphocyte ratio for the risk of AECOPD developing into pulmonary encephalopathy [J]. Journal of Dalian Medical University, 2022, 44(6): 511-514.
[6] 张力英, 李琦, 王苹, 等. 慢性阻塞性肺疾病合并肺性脑病的临床特征及其危险因素分析[J].中华肺部疾病杂志(电子版), 2021, 14(4): 530-532.
[7] Chen D, Zhao J, Ma M, et al. Dynamic nomogram for predicting acute kidney injury in patients with community-acquired pneumonia[J]. BMJ Open Respir Res, 2023, 10(1): e001495. doi: 10.1136/bmjresp-2022-001495
[8] Wu CW, Li SS, Xu E. Predictors of rapid eye movement sleep behavior disorder in patients with Parkinsons disease based on random forest and decision tree[J]. PLoS One, 2022, 17(6): e0269392. doi: 10.1371/journal.pone.0269392
[9] 中华医学会呼吸病学分会慢性阻塞性肺疾病学组, 中国医师协会呼吸医师分会慢性阻塞性肺疾病工作委员会. 慢性阻塞性肺疾病诊治指南(2021年修订版)[J]. 中华结核和呼吸杂志, 2021, 44(3): 170-205. Chronic Obstructive Pulmonary Disease Group of Chinese Thoracic Society, Chronic Obstructive Pulmonary Disease Committee of Chinese Association of Chest Physician. Guidelines for the diagnosis and management of chronic obstructive pulmonary disease(revised version 2021)[J]. Chinese Journal of Tuberculosis and Respiratory Diseases, 2021, 44(3): 170-205.
[10] 叶任高, 陆再英. 内科学[M]. 6版, 北京: 人民卫生出版, 2004: 137.
[11] 张红波, 蒋亚林, 代慧. 尼可刹米与洛贝林联合低分子肝素对AECOPD合并肺性脑病患者微循环及心肺功能的影响[J]. 医学综述, 2022, 28(16): 3314-3318.
[12] Dobric A, De Luca SN, Spencer SJ, et al. Novel pharmacological strategies to treat cognitive dysfunction in chronic obstructive pulmonary disease[J]. Pharmacol Ther, 2022, 233(1): 108017. doi: 10.1016/j.pharmthera.2021.108017
[13] Xia J, Yang H, Zhan Q, et al. High-flow nasal cannula may prolong the length of hospital stay in patients with hypercapnic acute COPD exacerbation[J]. Respir Med, 2023, 220(1): 107465. doi: 10.1016/j.rmed.2023.107465
[14] Hoiland RL, Robba C, Menon DK, et al. Clinical targeting of the cerebral oxygen cascade to improve brain oxygenation in patients with hypoxic-ischaemic brain injury after cardiac arrest[J]. Intensive Care Med, 2023, 49(9): 1062-1078.
[15] Shobatake R, Ota H, Takahashi N, et al. The impact of intermittent hypoxia on metabolism and cognition[J]. Int J Mol Sci, 2022, 23(21): 12957.doi: 10.3390/ijms232112957
[16] Samareh Fekri M, Nagibzadeh Tahami A, Arabnejad F. Relation of cognitive impairment with number of acute exacerbations and serum level of VEGF among COPD patients[J]. Tanaffos, 2023, 22(3): 311-316.
[17] Deep A, Behera PR, Subhankar S, et al. Serum electrolytes in patients presenting with acute exacerbation of chronic obstructive pulmonary disease(COPD)and their comparison with stable COPD patients[J]. Cureus, 2023, 15(4): e38080. doi: 10.7759/cureus.38080
[18] Hung SY, Liu WJ, Wu PC, et al. Noninvasive ventilator prevents intubation in thyrotoxic hypokalemic periodic paralysis-associated respiratory failure: a case report and literature review[J]. Tzu Chi Med J, 2021, 33(1): 91-95.
[19] 穆朝娟, 贾继超, 刘炳伦, 等. 精神障碍谵妄患者的危险因素分析[J]. 精神医学杂志, 2020, 33(3): 221-223.
[20] Munuswamy R, De Brandt J, Burtin C, et al. Monomeric CRP is elevated in patients with COPD compared to non-COPD control persons[J]. J Inflamm Res, 2021, 14(1): 4503-4507.
[21] Al-Baradie RS, Abdel-Hadi A, Ahmad F, et al. Association of monomeric C-reactive protein(m-CRP)with hypothalamic neurons after CRP hippo-campal administration in a model of dementia[J]. Eur Rev Med Pharmacol Sci, 2022, 26(23): 8713-8718.
[22] Orsolini L, Pompili S, Volpe U. C-reactive protein(CRP): apotent inflammation biomarker in psychiatric disorders[J]. Adv Exp Med Biol, 2023, 1411(1):135-160.
[23] Wu H, Ding P, Wu J, et al. A new online dynamic nomogram: construction and validation of a predictive model for distant metastasis risk and prognosis in patients with gastrointestinal stromal tumors[J]. J Gastrointest Surg, 2023, 27(7):1429-1444.
[24] Shi G, Liu G, Gao Q, et al. A random forest algorithm-based prediction model for moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia[J]. BMC Anesthesiol, 2023, 23(1):361. doi: 10.1186/s12871-023-02328-1
[25] Zhou Z, Chen C, Sun M, et al. A decision tree model to predict liver cirrhosis in hepatocellular carcinoma patients: a retrospective study[J]. PeerJ, 2023, 11(1): e15950. doi: 10.7717/peerj.15950
[1] 董雪,赵霞,程子捷,韩毅. 左西孟旦和米力农治疗重症心力衰竭合并肾损伤患者711例的药物经济学评价[J]. 山东大学学报 (医学版), 2022, 60(4): 91-98.
[2] 代晓宇,路媛,王志恒,李明卓,司书成,李吉庆, 井明, 薛付忠. 血常规数据判别骨髓增生异常综合征和急性髓样白血病的应用价值[J]. 山东大学学报 (医学版), 2020, 58(1): 20-25.
[3] 公晓云,申小涛,徐静,张涛,朱正江,薛付忠. 代谢组学数据正态性对疾病分类准确性的影响[J]. 山东大学学报（医学版）, 2016, 54(4): 89-93.
[4] 刘盈君, 张涛, 王璐, 刘佳, 常学润, 张敬悬, 薛付忠. 基于随机森林的精神分裂症血清代谢组学研究[J]. 山东大学学报（医学版）, 2015, 53(2): 92-96.
[5] 吴庆忠1，车峰远2，薛付忠1. 基于非平衡数据的癫痫发作预警模型研究[J]. 山东大学学报(医学版), 2012, 50(2): 141-.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2018 《山东大学学报(医学版)》编辑部　
地址:山东大学科技期刊社（济南市历城区山大南路27号山东大学中心校区明德楼B座721室） 电话: 0531-88366918 E-mail: xbyxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发