探讨原发性肝癌三维可视化诊治平台的构建及临床应用价值。

本前瞻性研究对象为2012年1月至2014年12月在南方医科大学珠江医院诊治的56例肝癌患者。所有患者均签署知情同意书，符合医学伦理学规定。其中男49例，女7例；平均年龄（47±7）岁。对患者行256层螺旋CT扫描，收集患者平扫期、动脉期、门静脉期和肝静脉期薄层CT数据。将数据导入医学图像三维可视化系统（MI-3DVS），分别对肝脏、肿瘤组织、门静脉系统、肝静脉系统、肝动脉系统及肝脏周围脏器行图像分割和三维可视化分析。然后行个体化肝脏分段、脉管及肝中叶肿瘤分型、个体化体积计算。通过模拟多种仿真手术进行手术规划，选择合理手术方式。对复杂性肝切除患者进行肝脏3D打印。

56例患者均完成三维可视化模型的构建，可清晰显示肝内各血管系统解剖形态、肿瘤部位及大小，进行肝脏个体化分段，脉管及肝中叶肿瘤分型。经个体化体积计算，患者全肝体积中位数为1 215（1 025～1 856）ml，肝肿瘤体积368（25～653）ml，虚拟切除肝体积478（125～854）ml。56例患者均完成肿瘤根治性切除，实际手术与术前手术规划一致，其中11例复杂性肝切除患者3D打印模型与术中所见完全一致。全部患者手术顺利，无发生围手术期死亡。6例患者术后出现胸腔积液，1例出现肝断面胆漏，经对症治疗后治愈。

原发性肝癌三维可视化诊治平台可实现术前精确诊断，术中精准操作，提高手术成功率。

To investigate the construction and clinical application of three-dimensional visualization platform in diagnosis and treatment of primary liver cancer.

Fifty-six patients with liver cancer diagnosed and treated in Zhujiang Hospital of Southern Medical University between January 2012 and December 2014 were enrolled in this prospective study. The informed consents of all patients were obtained and the local ethical committee approval had been received. Among the 56 patients, 49 were males and 7 were females with the average age of (47±7) years old. The 256-slice spiral CT scan was performed on the patients to collect plain scan data and thin-layer CT data of arterial phase, portal venous phase and hepatic venous phase. The data were imported into the Medical Image Three-dimensional Visualization System (MI-3DVS) to perform image segmentation and three-dimensional visualization analysis of the liver, tumor tissues, portal venous system, hepatic venous system, hepatic arterial system and organs around the liver. The individualized liver segmentation, classification of the vascular and middle lobe tumor was performed and individualized volume was calculated. The surgical planning was formulated through simulating multiple surgery and the rational surgical procedure was selected. For patients undergoing complex hepatectomy, three-dimensional liver model was printed.

The construction of three-dimensional visualization model of all 56 patients was completed, which could clearly display the anatomical morphology of each vascular system, the location and size of tumor, and perform individual liver segmentation as well as vascular, middle lobe tumor classification. According to individualized volume calculation, the median total liver volume was 1 215 (1 025-1 856) ml, the liver tumor volume was 368 (25-653) ml and the virtual removal liver volume was 478 (125-854) ml. All the 56 patients received radical resection of tumor. The actual operation was in accordant with the preoperative surgical plan. The 3D printing model of 11 patients undergoing complex hepatectomy was exactly the same with that observed during the operation. The operations of all patients were completed successfully and no death was observed during the perioperative period. Six patients developed pleural effusion and one developed bile leakage on the liver cutting surface after operation. These patients were cured after symptomatic treatments.

Three-dimensional visualization platform for diagnosis and treatment of primary liver cancer can realize the precise preoperative diagnosis and intraoperative manipulation, thus enhances the success rate of surgery.