Objective: To evaluate two radiological features, namely corona and mosaic in the prognosis in single hepatocellular carcinoma (HCC),as well as their value in guiding the selection between liver resection (LR) and transcatheter arterial chemoembolization (TACE). Methods: 275 patients with single HCC in > 5cm without extrahepatic metastasis treated by LR or TACE were included. In LR patients, the overall survival (OS) and time to progression (TTP) were compared between corona negative (corona-) vs. positive (corona+) and mosaic negative (mosaic-) vs. positive (mosaic+) respectively. Furthermore, by the combination of corona and mosaic, LR patients were divided to both negative (LR -/-), either positive (LR +/-) and both positive (LR +/+), their OS and TTP were compared with TACE group. Finally cox regression was applied to identify the independent factor for OS. Results: in survival plots, corona- had better OS and TTP than corona+. While mosaic- had better OS than mosaic-, but their TTP showed no statistical difference. In the whole population, LR -/- and LR +/- had similar OS, which were better than LR +/+ and TACE group. For TTP, LR -/- and LR +/- had better one than TACE, but there was no difference between LR -/- vs. LR +/-vs. LR +/+ and LR +/+ vs. TACE. In cox regression, the presence of corona/mosaic was independent factor for limited OS, both in LR patients and in the whole polulation. Conclusion: our results proved that for patients with single HCC > 5cm without extrahepatic metastasis, corona and mosaic were indications of limited LR efficacy. When both of the two features were identified, TACE could be used instead of LR without negative survival influence.