北京生物医学工程

基于LASSO方法智能诊断新冠肺炎重症与非重症

TheIntelligent diagnosis of severe and non-severe COVID-19based on LASSO methodZHONG Xiaoli,RENJinxia, XIAOFeng, XUHaibo

作者：钟小丽任金霞肖峰徐海波

单位：武汉大学中南医院医学影像科（武汉 430071）

关键词：新型冠状病毒肺炎；临床特征；特征提取；特征建模；鉴别诊断

分类号：R318

出版年·卷·期（页码）：2020·39·5（499-505）

摘要：

目的总结重症和非重症新型冠状病毒肺炎患者的临床特征，探讨基于特征分析智能诊断重症和非重症的可行性。方法回顾性分析确诊的112例新冠肺炎患者的临床资料，选取临床及实验室检查特征32个。首先分析所有特征的组间差异性，然后使用LASSO (least absolute shrinkage and selection operator)方法对存在组间差异的特征进行进一步的筛选及分类建模，最后采用ROC曲线、AUC值、灵敏度及特异性等指标对模型在训练集和测试集上进行评价。结果 32个临床特征中有23个特征存在组间差异，经过LASSO筛选后最终选定9个特征进行建模。模型的AUC（95%CI）、灵敏度、特异性：训练集0.962(0.927, 0.997)、0.912和0.909；测试集0.902(0.789, 1.000)、1.000和0.789。结论基于LASSO方法筛选的临床指标特征能够区分新冠肺炎非重症和重症，可为早期临床诊断治疗提供可靠依据。

Objective To summarize the clinical characteristics of patients with severe and non-severe COVID-19 ,and to explore the feasibility of intelligent diagnosis of severe and non-severe based on feature analysis. Methods As the clinical characteristics was one of the diagnostic criteria for suspected cases of new coronavirus, this study retrospectively analyzed the clinical data of 112 patients diagnosed with COVID-19 and extracted 32 clinical and laboratory features. First, we analyzed the differences between all the features. Then the least absolute shrinkage and selection operator (LASSO) method was used to further screen and construct a classification model for those features with differences between groups. Finally, we used indexes including the ROC curve, AUC value, sensitivity and specificity to evaluate the model on both the training and test sets. Results Among these 32 features, 23 features existed difference between groups and 9 features were finally selected for modeling after LASSO screening. The AUC (95% CI), sensitivity, and specificity of the model in the training set reached 0.962 (0.927, 0.997), 0.912, and 0.909, and in the test set these values were 0.902 (0.789, 1.000), 1.000, and 0.789. Conclusions The clinical features which are selected based on the LASSO method can distinguish between severe and non-severe COVID-19, which can provide a reliable basis for early clinical diagnosis and treatment.

参考文献：

[1]Gorbalenya AE, Baker SC, Baric RS, et al. Severe acute respiratory syndrome-related coronavirus: the species and its viruses–a statement of the Coronavirus Study Group[EB/OL]. BioRxiv, 2020. (2020-02-11). https://www.biorxiv.org/content/10.1101/2020.02.07.937862v1 [2]Lu R, Zhao X, Li J, et al. Genomic characterization and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding[J]. Lancet， 2020，39(10224): 565-574. [3]Otter JA, Donskey C, Yezli S, et al. Transmission of SARS and MERS coronaviruses and influenza virus in healthcare settings: the possible role of dry surface contamination[J]. Journal of Hospital Infection ，2016，92(3): 235-250. [4]Chan JF, Yuan S, Kok KH, et al. A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster[J]. Lancet ，2020，395(10223): 514-523. [5]中华人民共和国国家卫生健康委员会. 新型冠状病毒诊疗方案(试行第七版)[EB/OL]. [2020-03-03]. http://www.nhc.gov.cn/yzygj/s7653p/202003/46c9294a7dfe4cef80dc7f5912eb1989/files/ce3e6945832a438eaae415350a8ce964.pdf. [6]Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneumonia in China, 2019[J]. The New England Journal of Medicine， 2020，382(8):727-733. [7]Verity R, Okell LC, Dorigatti I, et al. Estimates of the severity of Coronavirus disease 2019: a model-based analysis[J]. Lancet Infect Dis,(2020-03-30),https://doi.org/10.1016/S1473-3099(20) 30243-7. [8]Tibshirani R. Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statistical Society, 1996, 58(1):267-288. [9]Tibshirani R. Regression shrinkage and selection via the lasso: a retrospective[J]. Journal of the Royal Statistical Society,2011, 73(3):273-282. [10]程克斌,魏明,沈虹,等.普通型和重型新型冠状病毒肺炎康复患者463例临床特征分析[J/OL].上海医学:1-15[2020-08-03].http://kns.cnki.net/kcms/detail/31.1366.r.20200312.1254.004.html. [11]Fu L,Wang B,Yuan T, et al. Clinical characteristics of coronavirus disease 2019 (COVID-19) in China: A systematic review and meta-analysis [J]. The Journal of Infection， 2020 ，80(6):656-665. [12]Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China[J]. The Lancet, 2020, 395（10223）：497-506. [13]Wang D, Hu B, Hu C, et al. Clinical Characteristics of 138 hospitalized patients with 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China [J]. JAMA, 2020， 323(11):1061-1069.. [14]Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study [J]. Lancet, 2020, 395(10 223):507-513. [15]戈艳蕾，李立群，王红阳，等. 老年重症社区获得性肺炎入住 ICU 患者血清 B 型钠尿肽、降钙素原、C反应蛋白水平及临床意义［J］. 中国老年学杂志，2018,38(10):2384-2385. [16]姚娜，王素娜，连建奇，等. 陕西地区新型冠状病毒肺炎合并肝损伤患者的临床特征与影响因素[J] .中华肝脏病杂志， 2020,28(3): 234-239. [17] Fan ZY , Chen LP , Li J , et al. Clinical features of COVID-19-related liver damage[J]. Clinical Gastroenterology and Hepatology, 2020, 18(7)：1-22. [18] 宋霞，陈涛，孙晶晶，等．PCT与IL-6联合检测对新冠病毒感染肺炎的临床价值探讨宋霞, 陈涛, 孙晶晶, et al. PCT与IL-6联合检测对新冠病毒感染肺炎的临床价值探讨[J/OL]. 国际检验医学杂志, 2020.（2020-03-10）. https://kns8.cnki.net/KCMS/detail/50.1176.R.20200309.1926.002.html

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录