Objective:To screen preoperative microvascular invasion (MVI)-related indicators in patients with hepatocellular carcinoma by machine learning, and to construct a predictive model for predicting MVI and evaluate it.Methods:The clinical data of hepatocellular carcinoma patients who underwent radical resection from January 2018 to March 2023 in General Hospital of Ningxia Medical University were retrospectively analyzed. A total of 437 patients were enrolled, including 325 males and 112 females, aged (56.3±13.6) years. The 437 patients were divided into a training set ( n=305) and a test set ( n=132) by computer-generated random numbers on a 7∶3 basis; the training set was used to construct the predictive model as well as to internally validate it by the five-fold cross-validation method, and the test set was used to externally validate the model. Two machine learning Boruta algorithm and LASSO regression, were used to screen MVI characteristic variables and construct multifactorial logistic regression prediction models. Receiver operating characteristic (ROC) curve, calibration curves, and decision curve were evaluated for predictive modeling, applying Shapley's additive explanatory analysis (SHAP) of the significance of key variables. Results:The intersection (5 variables) of 8 characteristic variables selected by Boruta algorithm and 8 variables selected by LASSO regression were selected: aspartate aminotransferase/lymphocyte ratio (ALR), tumor margin, intratumbral necrosis, tumor number and tumor maximum diameter, and the logistic regression model was constructed. The area under ROC curve for predicting the MVI were 0.77 (95% CI: 0.70-0.82) (training set), 0.76 (95% CI: 0.63-0.87) (validation set), and 0.84 (95% CI: 0.78-0.91) (test set). The prediction results of calibration curve logistic regression model were close to those of reagent, and the analysis of decision curve indicates that the model had good clinical application value. According to the mean absolute SHAP value, the order of importance was tumor margin, tumor maximum diameter, tumor number, ALR, and intratumoral necrosis. Conclusion:Tumor margin, tumor maximum diameter, tumor number, ALR and intratumoral necrosis were independent influencing factors for hepatocellular carcinoma associated with MVI, and the logistic regression model based on these factors was effective in predicting MVI.

Objective:To explore the application value of three-dimensional reconstruction in laparoscopic hepatectomy for liver cancer.Methods:Literature in Chinese or English about randomized controlled study (RCT), cohort study or case-control study of three-dimensional reconstruction combined with laparoscopic hepatectomy for liver tumors was retrieved in PubMed, Cochrane Library, Embase, SinoMed, CNKI, Wanfang and Chongqing VIP databases from the inception of database to February 2019. The Chinese and English searching terms included hepatoma, liver cancer, hepatectomy, laparoscope and three-dimensional reconstruction. Clinical data including perioperative condition and postoperative liver function were extracted for Meta-analysis.Results:8 articles of 534 patients were eventually included. Among them, 245 cases were assigned into the experimental group in which three-dimensional reconstruction was used and 289 cases in the control group. Meta-analysis showed that the operation time (MD=-36.48, 95%CI:-52.00 to -20.96, Z=4.61, P<0.05), intraoperative blood loss (MD=-109.36, 95%CI:-142.03 to -76.69, Z=6.56, P<0.05), incidence of postoperative complications (OR=0.42, 95%CI: 0.28 to 0.62, Z=4.38, P<0.05), length of postoperative hospital stay (MD=-3.31, 95%CI:-4.16 to -2.47, Z=7.66, P<0.05), postoperative ALT level (MD=-7.23, 95%CI: -8.39 to -6.07, Z=12.17, P<0.05), postoperative AST level (MD=-5.37, 95%CI:-8.59 to -2.14, Z=3.26, P<0.05) and postoperative TB level (MD=-2.84, 95%CI:-3.45 to -2.23, Z=9.09, P<0.05) significantly differed between two groups. Subgroup analysis showed that the heterogeneity of operation time, intraoperative blood loss and length of postoperative hospital stay was mainly caused by the difficulty of laparoscopic hepatectomy.Conclusions:Application of three-dimensional reconstruction in laparoscopic hepatectomy for liver cancer can improve the surgical safety and efficacy, shorten the operation time, reduce the intraoperative blood loss and lower the risk of postoperative complications.

目的:建立一项更易于临床操作、推广的关于原发性肝细胞癌伴微血管侵犯(MVI)术前早期诊断列线图预测模型，通过内部验证及外部验证对模型进行评估。方法:回顾性分析2017年1月至2020年12月期间宁夏医科大学总医院收治的294例肝细胞癌患者的临床资料。依据就诊时间分为两组：建模组( n＝231)和验证组( n＝63)。根据既往文献和相关临床经验且易于术前获取的原则，初步选取γ-谷氨酰基转移酶(GGT)、血小板计数/淋巴细胞计数比值(PLR)、纤维蛋白原/白蛋白比值(FAR)、淋巴细胞计数/单核细胞计数比值(LMR)、天门冬氨酸氨基转移酶/血小板计数比值(APRI)等指标进行考察，筛选确定肝细胞癌伴MVI的独立危险因素，并以此构建列线图预测模型，将验证组应用于模型进行外部验证。 结果:本研究共纳入294例患者，其中男性223例，女性71例，年龄(55.1±10.9)岁。建模组中MVI阳性95例，MVI阴性136例；验证组中MVI阳性38例，MVI阴性25例。多因素logistic回归分析结果显示，FAR>0.06、GGT>50 U/L、APRI>0.16、肿瘤长径>5 cm、LMR>3.57和PLR>98.75是肝细胞癌发生MVI的独立危险因素( P<0.05)。以此构建的列线图模型的实际预测结果与理想结果相接近，具有良好的预测性能，C指数均在0.71~0.90之间。运用决策曲线分析评估预测模型肝细胞癌术前MVI风险的临床净获益情况，结果显示，当净获益率>0时，预测模型阈值为4%~77%，表明该模型具有良好的临床应用价值。 结论:根据术前临床指标GGT、APRI、LMR、PLR、FAR及肿瘤长径构建列线图模型，可以简单、准确地对原发性肝癌伴MVI进行预测。

Objective:To evaluate the application of 3D visualization and 3D printing in the diagnosis and treatment of complex liver tumors.Methods:Clinical data of 30 patients with complex liver tumors admitted to General Hospital of Ningxia Medical University from January 2016 to October 2019 were retrospectively analyzed. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 20 patients were male and 10 female, aged 24-74 years with a median age of 51 years. Abdominal enhanced CT scan was performed before operation. 3D reconstruction was carried out using 3D visualization system. 1:1 physical model was constructed by 3D printer. Prior to operation, the relationship between tumors and surrounding vessels was analyzed through the 3D images and physical model. Simulated resection was performed and individualized diagnosis and treatment regimes were designed. Operative conditions and postoperative complications were observed. The simulated resected liver volume was compared with the actual resected liver volume by t test.Results:3D reconstructions of all the patients were completed, and 3D printed models of 10 cases were obtained. The intrahepatic vessels of grade Ⅲ and above were reconstructed. The relationship among the 3D anatomical morphology, location of tumors and its relation with the surrounding blood vessels could be explicitly displayed. 3 patients were with hepatic artery variations, 4 cases with hepatic vein variations and 2 cases with portal vein branches invaded by tumors. The mean liver volume was (1 779±325) ml, and the liver tumor volume was (572±238) ml. The consistency rate between preoperative simulated surgery and the actual surgery was 100%(30/30). The simulated resected liver volume was (896±405) ml, which did not differ from (815±270) ml of the actual resected liver volume (t=0.205, P>0.05). The operation time was (288±66) min, intraoperative blood loss was (722±390) ml, and the length of hospital stay was (20±6) d. Postoperative pleural effusion occurred in 10 patients and bile leakage in 3 cases. No perioperative death, massive bleeding, liver failure or other severe complications were observed.Conclusions:3D visualization and 3D printing is precise, safe and effective in preoperative design and resection of complex liver tumors.