Abstract: Objective: To analyze the computed tomography (CT) imaging features and evolution of different stages of coronavirus disease 2019 (COVID-19). Methods: A retrospective analysis was conducted on the CT images of 113 patients diagnosed with COVID-19 at Hebei Provincial People's Hospital between December 2022 and January 2023 to observe the trends of imaging changes. Results: All 113 patients were clinically diagnosed with COVID-19. Among these, 32, 41, 20, 15, and five patients underwent CT examination once, twice, three times, four times, and five times, respectively. A total of 259 CT examinations were performed in this group of 113 patients. Among them, 32 were early-stage (within 7 days of COVID-19 infection) examinations, 87 were progression-stage CT examinations, and 140 were recovery-stage CT examinations. Analysis of the imaging features of each CT examination of the patients was performed to identify the imaging features and evolution rules of COVID-19. Among the 32 examinations performed in 32patients with early-stage disease (within 7 days of COVID-19 infection), 26 cases showed ground-glass density shadows and six cases additionally showed solid nodules. In the progression stage (8 ~ 30 days after COVID-19 infection), among the 87 CT examinations in 74 patients, 62, 11, and one patient underwent examinations one, two, and three times, respectively. The 87 examinations revealed there 32 cases with ground-glass density shadows and 55 cases with additional solid nodules. In the recovery stage (12 ~ 57 days after COVID-19 infection), 89 patients underwent 140 CT examinations. Among these, 48, 32, eight, and one patient underwent CT examinations once, twice, three times, and four times, respectively. Among the 140 CT examinations, 48 cases showed ground-glass density shadows, while 90 cases additionally showed solid nodules. Moreover, 112 patients had multiple lesions in multiple lobes, with only one case having multiple lesions in a single lobe. Regarding the distributions, in the early stage, 13 cases had subpleural distributions and 19 cases had peribronchovascular and subpleural distributions. In the progression stage, 24 cases had subpleural distribution, and 63 cases had peribronchovascular and subpleural distributions. In the recovery stage, 48 cases had subpleural distribution, two cases showed complete absorption and improvement, and 90 cases had peribronchovascular and subpleural distribution, with 48 cases accompanied by reticular shadows. Regarding thickening of the interlobular septa and vessels within the lesions, in the early stage, 32 cases showed thickening of the interlobular septa and vessels within the lesions. In the progression stage, 85 cases showed thickening of the interlobular septa and 87 cases showed thickening of vessels within the lesions. In the recovery stage, five cases showed thickening of the interlobular septa, one case showed thickening of vessels within the lesions, and 48 cases were accompanied by linear shadows. Finally, in the early stage, one case showed bronchial gas inflation. In the progression stage, six cases showed pleural effusion, six cases showed bronchial inflation, and three cases showed pulmonary emphysema. In the recovery stage, two cases showed bronchial inflation and one case showed pleural effusion. Conclusion: The characteristics of CT images differed in patients with new coronary pneumonia at different times. Understanding this evolution is important to guide clinical treatment.