人民卫生出版社系列期刊
ISSN 2096-2738 CN 11-9370/R

中国科技核心期刊(中国科技论文统计源期刊)
2020《中国学术期刊影响因子年报》统计源期刊
美国化学文摘社(CAS)数据库收录期刊
日本科学技术振兴机构(JST)数据库收录期刊

新发传染病电子杂志 ›› 2021, Vol. 6 ›› Issue (3): 172-176.doi: 10.19871/j.cnki.xfcrbzz.2021.03.002

• 论著 • 上一篇    下一篇

白细胞介素-6与新型冠状病毒B.1.617.2变异株引起肺损伤相关性的人工智能定量分析

甘清鑫1, 刘晋新1, 杨彦鸿1, 胡天丽1, 张烈光1, 黄德扬1, 李美瑜2   

  1. 1.广州医科大学附属市八医院放射科,广州 510440;
    2.广州医科大学附属市八医院心血管内科,广州 510440
  • 收稿日期:2021-07-07 出版日期:2021-08-31 发布日期:2021-10-11
  • 通讯作者: 李美瑜,Email:gz8hlmy@126.com
  • 基金资助:
    广东省中医药局中医药科研项目(20211303)

Artificial intelligence quantitative analysis reveals the effect of interleukin-6 injury in 2019 novel coronavirus B.1.617.2

Gan Qingxin1, Liu Jinxin1, Yang Yanhong1, Hu Tianli1, Zhang Lieguang1, Huang Deyang1, Li Meiyu2   

  1. 1. Department of Radiology, Guangzhou Eighth People's Hospital, Guangzhou Medical University, Guangzhou 510440, China;
    2. Cardiovascular Division,Guangzhou Eighth People's hospital,Guangzhou Medical University, Guangzhou 510440, China
  • Received:2021-07-07 Online:2021-08-31 Published:2021-10-11

摘要: 目的 探讨白细胞介素-6(IL-6)与新型冠状病毒(2019-nCoV)B.1.617.2变异株引起肺损伤的相关性,提示临床应及时检测IL-6的水平和提前干预,以期降低重症肺炎发生的风险。方法 收集2021年5月21日至2021年6月21日广州医科大学附属市八医院收治的2019-nCoV B.1.617.2变异株感染患者123例为研究对象进行回顾性研究,轻症患者105例,重症患者18例。比较两组患者的临床特点、炎性指标水平、IL-6及人工智能(AI)测量肺损伤程度。结果 2019-nCoV B.1.617.2变异株感染的重症患者的年龄均值、血清降钙素原(PCT)峰值、IL-6峰值、AI测量肺损伤峰值占比均高于轻症组(P<0.05),重症组患者血清IL-6峰值均值为(197.28±105.70)pg/ml,AI测量肺损伤峰值占比均值(36.21±14.91)%,肺损伤以双侧肺部多发磨玻璃样变和实变为主要影像学特征。结论 IL-6高表达是2019-nCoV B.1.617.2变异株引起肺损伤的潜在的免疫风险因素,应用AI定量分析有助于判断肺炎的严重程度,提前干预,以期降低患者发展为重症肺炎的风险。

关键词: 新型冠状病毒, 病毒变异株, 肺炎, 白细胞介素-6, 人工智能

Abstract: Objective To investigate the correlation of lung injury caused by interleukin-6 (IL-6) in 2019 novel coronavirus (2019-nCoV) B.1.617.2. It is suggested that the level of IL-6 should be detected in time and intervention should be carried out in advance in order to reduce the risk of severe pneumonia. Methods A retrospective study was conducted to collect 123 patients infected with 2019-nCoV B.1.617.2 from the Guangzhou Eighth People's Hospital, Guangzhou Medical University, in May 21 to June 21, 2021. There were 105 mild cases and 18 severe cases. The clinical characteristics, inflammatory index levels, IL-6 and artificial intelligence(AI) lung injury degree of the two groups were analyzed. Results The mean value of age, peak value of procalcitonin (PCT), peak value of serum IL-6 and peak value of pulmonary injury measured by AI in patients with 2019-nCoV B.1.617.2 strain were higher than those in patients with mild disease group (P<0.05). The mean value of serum IL-6 in patients with severe disease group was (197.28±105.70)pg/ml. The peak value of lung injury measured by AI accounted for (36.21±14.91)%, and the main imaging features of lung injury were multiple ground glass opacity and consolidation in bilateral lungs. Conclusion High expression of IL-6 is a potential immune risk factor for lung injury in patients with 2019-nCoV B.1.167.2 strain. AI quantitative analysis helps to determine the severity of pneumonia in patients, intervention as early as possible. In order to reduce the risk of severe pneumonia.

Key words: 2019 novel coronavirus, Virus variant, Pneumonia, Interleukin-6, Artificial intelligence