Objective To evaluate clinical and epidemiological characteristics of patients who were confirmed with coronavirus disease 2019 (COVID-19) in Gansu Province. Methods Clinical data of 91 patients who were confirmed with COVID-19 in Gansu Province between January 23 and February 20, 2020 were collected, epidemiology, clinical manifestations and imaging features were analyzed. Results 91 patients with COVID-19 were distributed in 11 prefectures and cities in the whole province, and 13 were clustering cases, peak interval of occurrence of COVID-19 in Gansu Province was between January 25 and February 10, 2020. Clear epidemiological history was found in 62 cases, 10 patients (14.1%) were from Hubei epidemic area, 52 cases (73.2%) were clustering onset due to close contact with COVID-19 patients. 76 cases were mild type and common type, 15 were severe type and critically severe type; by the end of February 20, 2020, 65 cases (71.4%) were cured, 24 (26.4%) were under treatment, and 2 (2.2%) died. Among 71 cases with complete clinical and imaging data, 32 (45.1%) were male, with a median age of 45 (1-94) years and a median incubation period of 6 (1-13) days. The main clinical symptoms included:fever (n=60, 84.5%), cough (n=65, 91.5%), expectoration (n=23, 32.4%), fatigue (n=25, 35.2%), and shortness of breath (n=10, 14.1%). Characteristic changes in CT image included:ground-glass opacity and/or patchy opacity (n=65, 91.5%), air bronchogram and/or consolidation opacity (n=18, 25.4%), halo sign or reversed halo sign(n=15, 21.1%), 20 patients (28.2%) had combined CT image changes, including ground-glass opacity and/or patchy opacity combined with air bronchogram and/or consolidation opacity (n=11, 15.15%), and ground-glass opacity and/or patchy opacity combined with halo sign or reversed halo sign(n=9, 12.7%). Conclusion Most cases of COVID-19 in Gansu Province have clear epidemiological characteristics, mainly are imported and clustered cases, there is no gender difference in incidence of the disease, people are generally susceptible, imaging features of chest CT are conductive to clinical diagnosis.