Objective To explore the applicability of three-dimensional visualization technology in the planned hepatectomy of type Ⅲ and Ⅳ hilar cholangiocarcinoma. Methods The clinical data of thirty-six patients with type Ⅲ and Ⅳ hilar cholangiocarcinoma treated between January 2015 and December 2017 in the Department of Hepatobiliary Surgery was retrospectively analyzed. Three-dimensional visualization technology was used to reconstruct the 320 slice spiral 2D CT image of patients. By observing the location and size of the tumors and finding out about the relationship between the tumors and the peripheral portal vein, hepatic artery, hepatic vein and intrahepatic bile duct, the resectability of the tumors was assessed. Planned hepatectomy was proposed if the ratio of the residual liver volume was less than 40%. The volume of the residual liver was regenerated via portal vein embolization and biliary drainage. After 2 or 3 weeks, 3D visual reconstruction was performed a second time. Once the ratio of the residual liver volume was more than 40%, radical hepatectomy was carried out as planned. The difference between simulated surgery and actual surgery was studied. Results Three-dimensional visual reconstruction technology was used to display the relationship between the tumors and their surrounding tissues. Bismuth-Corlette classification suggested that there were 16 cases of ⅢA, 8 cases of ⅢB and 12 cases of Ⅳ. The average volume was (1386±146) ml for whole livers,(76±22) ml for tumors,(896±168) ml for the pre-resected liver, and (490±172) ml for residual livers. The ratio of the residual liver volume (34.5%±3.6%) was measured. When the ratio of the residual liver volume exceeded 40% after embolization of the portal vein on the healthy side of biliary drainage, radical resection of hilar cholangiocarcinoma was performed. The accuracy of three-dimensional reconstruction was 100%. The anatomic relationship of the reconstruction model was consistent with what was observed in the course of surgery. There was no significant difference between the volume of the pre-resected liver and the actual volume of the postoperative liver (t=1.148). Conclusions 3D visualization technology can accurately assess hilar cholangiocarcinoma before operation, optimize the treatment plan and improve the R0 resection rate.