Primary liver cancer is the most common malignant tumor of the liver. Radical surgery is the preferred treatment, but its 5-year recurrence rate is as high as 70%, which is often associated with incomplete surgical resection and residual tumor. Since the 21st century, with the rapid development of surgery and new science and technology, the diagnosis and treatment of liver diseases have been closely linked with imaging diagnostics, biomedical engineering, molecular imaging technology, computer science and other interdisciplinary subjects. Three-dimensional (3D) visualization technology is one of the emerging auxiliary diagnosis and treatment methods. It plays an important role in accurate diagnosis, surgical planning and surgical navigation of primary liver cancer, and can effectively improve the success rate of surgery and reduce the incidence of postoperative complications. Its diagnostic and treatment value in primary liver cancer is mainly reflected in :(1) to clarify the spatial location relationship of abdominal space occupying lesions, understand the relationship between space occupying lesions and surrounding tissues, important blood vessels and cavities, judge tumor resectable, and make preoperative planning for tumor classification and surgical resection scope. (2) Calculation of individual liver segmentation and liver volume based on the topological relationship of blood flow 4. 3D simulation software was used to automatically calculate the resectioned liver volume and residual liver volume with statistical correlation, which was helpful to judge the probability of liver failure after liver cancer. (3) Variation of hepatic artery, portal vein and bile duct can be found, and the variation of duct can be evaluated by three-dimensional model, including whether there is contact, length of contact, whether there is stenosis in lumen, etc., especially the resectable ability of tumor patients can be evaluated by 3D and real-time dynamic navigation during surgery.

Intervention of the experimental group: based on high-quality thin-slice CT image reconstruction 3d visualization model to guide preoperative planning and surgery. The surgical process and perioperative management are as follows: All patients in this study were operated by the medical team whose main member was the medical team leader, and all patients were treated with anatomic hepatectomy. The main points of anatomic hepatectomy are summarized as follows :(1) determine and mark the boundary of liver segment according to ischemia line or staining. (2) The postoperative liver section should fully expose the iconic veins and vessels. (3) The initial part of Glisson's pedicle of the target liver segment was severed. Tissue was taken from the surgical margins of all patients and sent to frozen sections for examination. Routine pathological examinations were performed on all resected liver after surgery. Postoperative patients were treated with relevant perioperative symptomatic treatment, and blood routine examination, liver function and coagulation function were detected on the 1st, 3rd and 5th day after surgery.

Intervention in the control group: preoperative planning and operation were guided based on high-quality thin-slice CT images. Perioperative management same as above.

Inclusion Criteria:

• Patients with primary liver cancer clinically diagnosed preoperatively (tumor ≥3cm to ≤10cm, diagnostic criteria: Primary liver cancer Diagnosis and Treatment Code 2019 edition);
• No tumor thrombi formation and distant metastasis were found in the imaging data;
• Child-Pugh grading standard of liver function was GRADE A or B;
• 18-66 years old;
• Complete clinical case data;
• all patients underwent surgical treatment;
• Voluntarily participate in the study and sign the informed consent.

Exclusion Criteria:

• There are basic diseases that cannot tolerate surgery (such as severe cardiopulmonary cerebral renal insufficiency);
• Preoperative imaging examination found cancer thrombus in main portal vein and branches, common hepatic vein and branches, main hepatic vein and branches and inferior vena cava;
• planned pregnancy, unplanned pregnancy and pregnancy;
• Preoperative child-Pugh grading standard of liver function was Grade C.
• Disease researchers that the investigator considers inappropriate to participate in this clinical trial.

A single blind design was used in this study. Patients were not informed about whether to use 3D visualization technology, and the subjects were blinded. Random numbers used for grouping are generated by third-party statisticians using SPSS. After the patients were enrolled, the nurses provided the surgeons with random numbers and groups of the patients in identical, opaque, sealed envelopes. The 3 researchers who will not participate in the operation will be independently responsible for recording the preoperative 3D visualization or CT analysis results, surgical photos and information records, postoperative complications and prognosis follow-up of the patients. The statisticians on the team who analyzed the data did not know how the patients were grouped.