Guo Xing, Li Yanrong, Li Hua, Li Xueqin, Chang Xu, Bai Xuemei, Song Zhanghong, Li Junfeng, Li Kefeng
Department of Radiology, Heping Hospital Affiliated to Changzhi Medical College, Shanxi 046000, China.
Department of Pharmacy, Changzhi Medical College, Shanxi 046000, China.
Aging (Albany NY). 2020 Oct 21;12(20):19938-19944. doi: 10.18632/aging.104132.
COVID-19 shared many symptoms with seasonal flu, and community-acquired pneumonia (CAP) Since the responses to COVID-19 are dramatically different, this multicenter study aimed to develop and validate a multivariate model to accurately discriminate COVID-19 from influenza and CAP. Three independent cohorts from two hospitals (50 in discovery and internal validation sets, and 55 in the external validation cohorts) were included, and 12 variables such as symptoms, blood tests, first reverse transcription-polymerase chain reaction (RT-PCR) results, and chest CT images were collected. An integrated multi-feature model (RT-PCR, CT features, and blood lymphocyte percentage) established with random forest algorism showed the diagnostic accuracy of 92.0% (95% CI: 73.9 - 99.1) in the training set, and 96. 6% (95% CI: 79.6 - 99.9) in the internal validation cohort. The model also performed well in the external validation cohort with an area under the receiver operating characteristic curve of 0.93 (95% CI: 0.79 - 1.00), an F1 score of 0.80, and a Matthews correlation coefficient (MCC) of 0.76. In conclusion, the developed multivariate model based on machine learning techniques could be an efficient tool for COVID-19 screening in nonendemic regions with a high rate of influenza and CAP in the post-COVID-19 era.
新冠病毒病（COVID-19）与季节性流感及社区获得性肺炎（CAP）有许多共同症状，由于对 COVID-19 的应对措施截然不同，这项多中心研究旨在开发并验证一个多变量模型，以准确区分 COVID-19 与流感及 CAP。纳入了来自两家医院的三个独立队列（发现集和内部验证集各 50 例，外部验证队列 55 例），收集了症状、血液检测、首次逆转录聚合酶链反应（RT-PCR）结果及胸部 CT 图像等 12 个变量。用随机森林算法建立的综合多特征模型（RT-PCR、CT 特征及血液淋巴细胞百分比）在训练集中的诊断准确率为 92.0%（95%CI：73.9 - 99.1），在内部验证队列中为 96.6%（95%CI：79.6 - 99.9）。该模型在外部验证队列中也表现良好，受试者工作特征曲线下面积为 0.93（95%CI：0.79 - 1.00），F1 分数为 0.80，马修斯相关系数（MCC）为 0.76。总之，基于机器学习技术开发的多变量模型可能是 COVID-19 后时代流感和 CAP 高发的非流行地区 COVID-19 筛查的有效工具。
