基于深度学习的慢性阻塞性肺病的诊断模型研究

doi:10.3969/j.issn.0258-8021.2022.05.005

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2770 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要慢性阻塞性肺病(COPD)是一种常见的以持续气流受限为特征的慢性呼吸道疾病,具有很高的发病率和死亡率。目前临床上对COPD的诊断方式十分复杂,不仅耗时且有创或有辐射伤害,不适用于日常筛查。本研究设计了一种基于深度学习的COPD诊断模型。首先,将RespiratoryDatabase @ TR多媒体呼吸数据库中42位COPD患者的肺音数据和来自天津大学胸科医院的24位COPD患者以及37位健康受试者的临床采集肺音数据相结合,分别运用高通滤波器和基于集合经验模态分解(EEMD)及小波熵的去噪算法进行去噪处理,然后通过归一化、交叠剪切、数据扩增完成预处理过程;然后利用二阶谱分析技术提取肺音特征;最后,将特征输入到改进的19层卷积神经网络模型中,实现健康受试者与COPD患者的二分类。实验结果表明,所提出的模型能够有效诊断COPD,其准确度、敏感度、特异性、F1分数和Kappa系数分别达到了98.93%、98.47%、99.41%、98.95%和97.86%,且由于采用了双中心数据并进行了去噪处理,模型可靠性更高,具有重要的临床意义。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章

关键词 ：慢性阻塞性肺病, 肺音, 二阶谱, 卷积神经网络

Abstract：Chronic obstructive pulmonary disease (COPD) is a common chronic respiratory disease characterized by continuous airflow restriction, with high morbidity and mortality. At present, clinical diagnosis methods of COPD are very complex, not only time-consuming and invasive or radioactive, and not suitable for daily screening. Therefore, a COPD diagnosis model based on deep learning was designed in this study. Firstly, the lung sound from 42 COPD patients from RespiratoryDatabase@TR multimedia respiratory database were combined with the clinically collected lung sound from 24 COPD patients and 37 healthy subjects from Chest Hospital of Tianjin University, high-pass filter and denoising algorithm based on ensemble empirical mode decomposition (EEMD) and wavelet entropy was used for denoising. Secondly, the pre-processing process was completed through normalization, overlapping shear and data amplification. Thirdly, bispectrum analysis was used to extract the lung sound features. Finally, these features were input into an improved 19-layer convolutional neural network model to achieve the binary classification of healthy subjects and COPD patients. Experimental results showed that the proposed model could effectively diagnose COPD. The accuracy, sensitivity, specificity, F1 score, and Kappa score reached 98.93%, 98.47%, 99.41%, 98.95%, and 97.86%, respectively. Moreover, due to the use of bicentric data and denoising process, the model has higher reliability and is of important clinical significance.

Key words： chronic obstructive pulmonary disease lung sound bispectrum analysis convolutional neural network

收稿日期: 2021-09-14

PACS:

R318

基金资助:国家重点研发计划(2019YFC0119402);天津科技重大专项与工程资助项目(18ZXZNSY00240);天津大学研究生创新人才培养项目(YCX202001)

通讯作者: * E-mail: sunjinglai@tju.edu.cn; zxydoctor@163.com

作者简介: ^#中国生物医学工程学会会员

引用本文:

余辉, 赵婧, 仇兆禹, 刘冬怡, 陈振, 垢程翔, 孙敬来, 赵晓赟. 基于深度学习的慢性阻塞性肺病的诊断模型研究[J]. 中国生物医学工程学报, 2022, 41(5): 558-566.
Yu Hui,#,Zhao Jing,Qiu Zhaoyu,Liu Dongyi,Chen Zhen,Gou Chengxiang,Sun Jinglai*,Zhao Xiaoyun*. Diagnosis Model of Chronic Obstructive Pulmonary Disease Based on Deep Learning. Chinese Journal of Biomedical Engineering, 2022, 41(5): 558-566.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021.2022.05.005 或 http://cjbme.csbme.org/CN/Y2022/V41/I5/558

[1] 葛均波,徐永健,内科学(第8版) [M].北京:人民卫生出版社,2013:21-27.
[2] Wang Chen, Xu Jianying, Yang Lan, et al. Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China pulmonary health [CPH] study): a national cross-sectional study [J]. The Lancet, 2018, 391(10131): 1706-1717.
[3] Balkissoon R. Global initiative for chronic obstructive lung disease 2020 report and the journal of the COPD foundation special edition, moving to a new definition for COPD: “COPDGene® 2019” [J]. Chronic Obstructive Pulmonary Diseases, 2020, 7(1): 64.
[4] Patel AR, Patel AR, Singh S, et al. Global initiative for chronic obstructive lung disease: The changes made [J]. Cureus, 2019, 11(6): e4985-e4991.
[5] Kinney GL, Santorico SA,Young KA, et al. Identification of chronic obstructive pulmonary disease axes that predict all-cause mortality: the COPDgene study [J]. American Iournal of Epidemiology, 2018, 187(10): 2109-2116.
[6] Gershon AS, Hwee J, Croxford R, et al. Patient and physician factors associated with pulmonary function testing for COPD: a population study [J]. Chest, 2014, 145(2): 272-281.
[7] 陈仕锋, 黄敏於, 彭显如, 等. 肺音可以作为首诊慢阻肺严重程度的判断指标 [J]. 南方医科大学学报, 2020, 40(2): 177-182.
[8] Sarkar M, Bhardwaz R, Madabhavi I, et al. Physical signs in patients with chronic obstructive pulmonary disease [J]. Lung India, 2019, 36(1): 38-47.
[9] Cristina J, Ana O, Alda M. Computerized respiratory sounds: a comparison between patients with stable and exacerbated COPD [J]. The Clinical Respiratory Journal, 2017, 11(5): 612-620.
[10] Baroi S, McNamara RJ, McKenzie DK, et al. Advances in remote respiratory assessments for people with chronic obstructive pulmonary disease: A systematic review [J]. Telemedicine Journal and e-Health, 2018, 24(6): 415-424.
[11] Jácome C, Marques A. Computerized respiratory sounds: novel outcomes for pulmonary rehabilitation in COPD [J]. Respiratory Care, 2017, 62(2): 199-208.
[12] Yasuharu T, Seishirou M. Physical diagnosis of chronic obstructive pulmonary disease [J]. The Japanese Society of Internal Medicine, 2007, 46(23): 1885-1891.
[13] Pramono RXA, Bowyer S, Rodriguez-Villegas E. Automatic adventitious respiratory sound analysis: A systematic review [J]. PLoS ONE, 2017, 12(5): e0177926-e0177968.
[14] Wang Qian, Wang Hong, Wang Lutong, et al. Diagnosis of chronic obstructive pulmonary disease based on transfer learning [J]. IEEE Access, 2020, 8: 47370-47383.
[15] Ying Jun, Dutta J, Guo Ning, et al. Classification of exacerbation frequency in the COPDgene cohort using deep learning with deep belief networks [J]. IEEE Journal of Biomedical & Health Informatics, 2020, 24(6): 1805-1813.
[16] Morillo DS, León JA, Moreno SA. Computer-aided diagnosis of pneumonia in patients with chronic obstructive pulmonary disease [J]. Journal of American Medical Informatics Association, 2013, 20: e111-e117.
[17] Altan G, Kutlu Y, Pekmezci AO, et al. Deep learning with 3D-second order difference plot on respiratory sounds [J]. Biomedical Signal Processing & Control, 2018, 45(8): 58-69.
[18] Ahmet G. Computer-aided diagnosis system for chronic obstructive pulmonary disease using empirical wavelet transform on auscultation sounds[J]. The Computer Journal, 2021, 64: bxaa191.
[19] Altan G, Kutlu Y. Hessenberg elm autoencoder kernel for deep learning[J]. Journal of Engineering Technology and Applied Sciences, 2018, 3(2): 141-151.
[20] Altan G, Kutlu Y, Allahverdi N. Deep learning on computerized analysis of chronic obstructive pulmonary disease [J]. IEEE Journal of Biomedical and Health Informatics, 2020, 24(5): 1344-1350.
[21] Altan G, Kutlu Y, Gken A. Chronic obstructive pulmonary disease severity analysis using deep learning on multi-channel lung sounds [J]. Turkish Journal of Electrical Engineering and Computer Sciences, 2020, 28(5): 2979-2996.
[22] Altan G, Kutlu Y, Garbi Y, et al. Multimedia respiratory database (respiratorydatabase@tr): auscultation sounds and chest X-rays [J]. Natural and Engineering Sciences, 2017, 2(3): 59-72.
[23] Meslier N, Charbonneau G, RAc Ineux JL. Wheezes [J]. European Respiratory Journal, 1995, 8(11): 1942-1948.
[24] Wu Zhaohua, Huang NE. Ensemble empirical mode decomposition: a noise-assisted data analysis method [J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41.
[25] 张艺还, 王旭红, 郭良, 等. 基于归一化自相关函数与类小波软阈值法的GIS局放信号降噪方法研究 [J]. 高压电器, 2018, 54(3): 17-24.
[26] 高鸣蕾. 基于EMD的肺音信号去噪及特征提取方法研究 [D]. 镇江: 江苏大学, 2019.
[27] Chen Xianyao, Wang Meng, Zhang Yuanling, et al. Detecting signals from data with noise: theory and applications [J]. Journal of the Atmospheric Sciences, 2013, 70(5): 1489-1504.
[28] 崔东东, 张恒璟, 程鹏飞. 一种自相关函数绝对值均值变点的去噪方法 [J]. 测绘科学, 2019, 44(12): 42-49.
[29] 张樯, 周西峰, 王瑾, 等. 基于改进的EMD超声信号降噪方法研究 [J]. 南京邮电大学学报(自然科学版), 2016, 36(2): 49-55.
[30] 余发军, 周凤星. 基于EEMD和自相关函数特性的自适应降噪方法 [J]. 计算机应用研究, 2015(1): 206-209.
[31] 董利超, 郭兴明, 郑伊能. 基于CEEMD的心音信号小波包去噪算法研究 [J].振动与冲击, 2019, 38(9): 192-198,222.
[32] Alqudah AM, Alquran HH, Abuqasmieh I. Classification of heart sound short records using bispectrum analysis approach images and deep learning [J]. Network Modeling Analysis in Health Informatics and Bioinformatics, 2020, 9(12): 66-71.
[33] Khadra L, Al-Fahoum AS, Binajjaj S. A quantitative analysis approach for cardiac arrhythmia classification using higher order spectral techniques [J]. IEEE Transactions on Biomedical Engineering, 2005, 52(11): 1840-1845.
[34] 赵守国, 王素品, 邹原, 等. 肺音信号非高斯ARMA模型及双谱研究 [J]. 西安交通大学学报, 2001(4): 355-359.
[35] Alqudah AM. AOCT-NET: A convolutional network automated classification of multiclass retinal diseases using spectral-domain optical coherence tomography images [J]. Medical & Biological Engineering & Computing, 2019, 58(1): 41-53.