Objective To explore the application value of Flash chest CT scan protocol in the follow-up of severe coronavirus disease 2019 (COVID-19). Methods Twenty-eight patients with severe COVID-19 who were admitted to this hospital from February 1, 2020 to March 15, 2020 were included. The follow-up CT examinations used conventional chest CT scan protocol and Flash scan protocol respectively, the image quality, signal-to-noise ratio (SNR) and effective dose of these two scan protocols were compared. Kappa test was used to evaluate the consistency of image quality scores of readers; Mann-Whitney test was used for image quality comparison; paired t test is performed for signal-to-noise ratio and effective dose comparison. Results The scores of the two readers had good consistency (Kappa=0.62, P<0.01 and Kappa=0.67, P<0.01). There was no statistically significant difference in image quality scores (Z=–0.275, P=0.783 and Z=–0.212, P=0.832). The signal-to-noise ratio of Flash protocol was statistically higher than that of conventional protocol (2.30±0.30 vs. 1.76±0.25, P<0.01). The effective dose of Flash protocol was 66.8% lower than that conventional protocol [(1.43±0.32) mSv vs. (4.31±1.36) mSv]. Conclusion Compared with conventional scan protocol, the Flash scan protocol does not require contact with COVID-19 patients, and effectively reduces the radiation dose and improves image quality of severe COVID-19 patients.