| Harrison X. Bai, Ben Hsieh, Zeng Xiong, Kasey Halsey, Ji Whae Choi, Thi My Linh Tran, Ian Pan, Lin-Bo Shi, Dong-Cui Wang, Ji Mei, Xiao-Long Jiang, Qiu-Hua Zeng, Thomas K. Egglin, Ping-Feng Hu, Saurabh Agarwal, Fangfang Xie, Sha Li, Terrance Healey, Michael K. Atalay, Wei-Hua Liao
This study evaluates the performance of radiologists in differentiating COVID-19 from viral pneumonia on chest CT scans. The research involved 219 Chinese patients with confirmed COVID-19 and 205 non-COVID-19 patients from the United States. Three Chinese radiologists and four U.S. radiologists reviewed the chest CT scans, which were blindly assessed to avoid bias. The results showed that the Chinese radiologists had a sensitivity of 72%, 72%, and 94%, and a specificity of 94%, 88%, and 24%. The U.S. radiologists had a sensitivity of 93%, 83%, 73%, and 73%, and a specificity of 100%, 93%, 93%, and 100%. Key discriminating features for COVID-19 pneumonia included peripheral distribution (80% vs. 57%, p<0.001), ground-glass opacity (91% vs. 68%, p<0.001), and vascular thickening (58% vs. 22%, p<0.001). The study concludes that radiologists can distinguish COVID-19 from viral pneumonia with high specificity but moderate sensitivity, highlighting the importance of combining chest CT findings with RT-PCR results for accurate diagnosis.This study evaluates the performance of radiologists in differentiating COVID-19 from viral pneumonia on chest CT scans. The research involved 219 Chinese patients with confirmed COVID-19 and 205 non-COVID-19 patients from the United States. Three Chinese radiologists and four U.S. radiologists reviewed the chest CT scans, which were blindly assessed to avoid bias. The results showed that the Chinese radiologists had a sensitivity of 72%, 72%, and 94%, and a specificity of 94%, 88%, and 24%. The U.S. radiologists had a sensitivity of 93%, 83%, 73%, and 73%, and a specificity of 100%, 93%, 93%, and 100%. Key discriminating features for COVID-19 pneumonia included peripheral distribution (80% vs. 57%, p<0.001), ground-glass opacity (91% vs. 68%, p<0.001), and vascular thickening (58% vs. 22%, p<0.001). The study concludes that radiologists can distinguish COVID-19 from viral pneumonia with high specificity but moderate sensitivity, highlighting the importance of combining chest CT findings with RT-PCR results for accurate diagnosis.