Prediction of early recurrence of hepatocellular carcinoma after liver transplantation based on CT radiomics model
ZHAO Jingwei1,2, CHEN Xiaoxia2, GUO Xiaodong3, HE Xucheng2, HAN Wenjuan2, YAO Dingming2, LIU Ting2, WANG Guisheng2
1. Postgraduate Training Base, The Third Medical Center of Chinese PLA General Hospital, Jinzhou Medical University, Beijing 100039, China; 2. Department of CT, The Third Medical Center of Chinese PLA General Hospital, Beijing 100039, China; 3. The Fifth Medical Center of Chinese PLA General Hospital,Beijing 100039,China
摘要目的 探索基于CT影像组学技术构建的模型在预测肝细胞癌患者肝移植术后早期复发的价值。方法 回顾性分析接受肝移植治疗的131例肝癌患者,随机分为训练组(92例)和验证组(39例),术后定期随访,了解是否发生早期复发。通过逐层勾画肿瘤边缘对肿瘤进行三维分割并进行特征提取,共提取1218个影像组学特征。具有潜在预测价值特征的筛选选用LASSO算法。基于筛选出的特征,logistic回归应用于肝移植术后预测模型的构建。通过曲线下面积(area under the curve, AUC)对模型预测患者是否会早期复发的效能进行评价。结果 筛选出8个具有潜在预测价值的特征,预测模型在训练组中AUC为0.828,敏感度、特异度分别为82.4%、74.7%;在验证组中AUC为0.856,敏感度、特异度分别为77.8%、86.7%。结论 术前增强CT影像组学技术构建的模型,对预测肝癌肝移植术后复发具有一定价值。
Abstract：Objective To explore the value of CT radiomics model in predicting early recurrence in patients with hepatocellular carcinoma (HCC) after liver transplantation.Methods A retrospective review of 131 patients with HCC who underwent liver transplantation was performed. And the whole dataset were randomly split into two datasets,a training dataset(92 cases) and a validation dataset(39 cases). Early recurrence was confirmed by postoperative regular follow-up. Three-dimensional tumor segmentation is completed by delineating the tumor margins slice by slice and 1218 features were gengerated. To select features with potential predictive value,The LASSO algorithm was applied. By applying logistic regression,the model for predicting early recurrence after liver transplantation were developed based on the selected features. The performance of the model to predict whether the patients would experience early recurrence was evaluated by the area under the curve (AUC).Results The selected features with potential predictive value were 8.The prediction model showed AUC=0.828 , sensitivity=82.4% , specificity=74.7% in training dataset and AUC=0.856, sensitivity=77.8%, specificity=86.7% in validation dataset.Conclusions The radiomic model based on preoperative enhanced CT images has predictive value for early recurrence of HCC patients after liver transplantation.
赵经纬, 陈晓霞, 郭晓东, 何绪成, 韩文娟, 姚鼎铭, 刘婷, 王贵生. CT影像组学模型预测肝移植术后肝癌早期复发的价值[J]. 武警医学, 2021, 32(5): 399-402.
ZHAO Jingwei, CHEN Xiaoxia, GUO Xiaodong, HE Xucheng, HAN Wenjuan, YAO Dingming, LIU Ting, WANG Guisheng. Prediction of early recurrence of hepatocellular carcinoma after liver transplantation based on CT radiomics model. Med. J. Chin. Peop. Armed Poli. Forc., 2021, 32(5): 399-402.
Taketomi A,Fukuhara T,Morit K, et al. Improved results of a surgical resection for the recurrence of hepatocellular carcinoma after living donor liver transplantation.[J].Ann Surg Oncol,2010,17(9):2283-2289.
Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis[J]. N Engl J Med, 1996, 334(11):693-700.
[9]
Reddy M S, Manas D M. Accuracy of staging as a predictor for recurrence after liver transplantation for hepatocellular carcinoma[J]. Transplantation, 2007, 83(3):367-368.
Ning P, Gao F, Hai J, et al. Application of CT radiomics in prediction of early recurrence in hepatocellular carcinoma[J]. Abdom Radiol (NY), 2020,45(9):64-72.
[12]
Zhu H B, Zheng Z Y, Zhao H, et al. Radiomics-based nomogram using CT imaging for noninvasive preoperative prediction of early recurrence in patients with hepatocellular carcinoma[J]. Diagn Interv Radiol, 2020, 26(5): 411-419.
[13]
Ji G W, Zhu F P, Xu Q, et al. Radiomic features at contrast-enhanced CT predict recurrence in early stage hepatocellular carcinoma: a multi-institutional study[J]. Radiology, 2020, 294(3): 568-579.
2021, Vol. 32 
