河北医科大学学报 ›› 2024, Vol. 45 ›› Issue (7): 771-778.doi: 10.3969/j.issn.1007-3205.2024.07.006

• • 上一篇    下一篇

多参数MR为基础的CNN模型预测肝细胞性肝癌的微血管侵犯

  

  1. 1.河北省保定市第二中心医院放射科，河北 涿州 072750；2. 河北省保定市第二中心医院消化科，河北 涿州 072750；
    3.河北省保定市第二中心医院急诊科，河北 涿州 072750；4.河北省保定市第二中心医院CT室，
    河北 涿州 072750；5.河北省保定市第二中心医院五官科，河北 涿州 072750

  • 出版日期:2024-07-25 发布日期:2024-07-18
  • 作者简介:王谦(1982-)，男，河北唐县人，河北省保定市第二中心医院主治医师，医学学士，从事胸腹部影像学诊断研究。
  • 基金资助:
    保定市科技计划项目（2341ZF037）

CNN model based on multi-parameter MR in predicting microvascular invasion of HCC

  1. 1.Department of Radiology, the Second Central Hospital of Baoding City, Hebei Province, Zhuozhou 
    072750, China; 2.Department of Gastroenterology, the Second Central Hospital of Baoding City, 
    Hebei Province, Zhuozhou 072750 , China; 3.Department of Emergency, the Second Central 
    Hospital of Baoding City, Hebei Province, Zhuozhou 072750, China; 4.Department of CT, 
    the Second Central Hospital of Baoding City, Hebei Province, Zhuozhou 072750, China; 
    5.Department of Otorhinolaryngology, the Second Central Hospital of 
    Baoding City, Hebei Province, Zhuozhou 072750, China

  • Online:2024-07-25 Published:2024-07-18

HTML

PDF (PC)

98

摘要： 目的 采用基于多参数磁共振序列的卷积神经网络（convolutional neural network，CNN）联合传统影像组学标签及临床指标，术前预测肝细胞性肝癌（hepatocellular carcinoma，HCC）患者的微血管侵犯（microvascular invasion，MVI）。
方法 选择经病理确诊的HCC患者275例纳入本研究。将数据集随机分为训练集（n=192）和测试集（n=83）。应用CNN技术，融合二维多参数磁共振肿瘤图像、三维肿瘤的传统影像组学特征标签及临床指标，开发一种HCC的MVI预测分类器。应用受试者工作特征曲线（receiver operating characteristic,ROC），比较混合模型（ModelCom）与卷集神经网络模型（ModelD）、影像组学模型（ModelR）和临床模型（ModelC）的诊断效能。
结果 ModelD在训练集和测试集中的AUC分别为0.914和0.842，优于ModelC（训练集：P＜0.001；测试集:P=0.032）和ModelR（训练集：P＜0.001；测试集:P=0.044）。ModelCom在训练集和测试集中的AUC分别为0.951和0.881，在训练集中优于ModelD（P=0.012），在测试集中差异无统计学意义（P=0.157）。校准曲线显示出了ModelCom具有良好的拟合优度（hosmer-lemeshow test,训练集P=0.402，测试集P=0.689）。决策曲线分析提示ModelCom鉴别MVI阳性和MVI阴性的净获益高于其他模型。
结论 CNN为基础的混合模型够准确预测HCC的MVI状态。


关键词: 肝肿瘤, 卷积神经网络, 影像组学

Abstract: Objective To explore convolutional neural network (CNN) based on multi-parameter magnetic resonance (MR) sequences, combined with traditional radiomics signature and clinical indicators, in predicting microvascular invasion (MVI) of hepatocellular carcinoma (HCC) preoperatively.  
Methods Two hundred and seventy-five patients with pathologically confirmed HCC were enrolled in this study. The data set was randomly divided into a training set (n=192) and a test set (n=83). A MVI predictive classifier for HCC was developed by using CNN technique, which fused 2D multi-parameter MR tumor images, 3D traditional radiomics signature and clinical indicators. Using the receiver operating characteristic (ROC) curve, the performance of combined model (ModelCom), CNN model (ModelD), radiomics model (ModelR) and clinical model (ModelC) were compared. 
Results The area under the ROC curve (AUC) of ModelD was 0.914 in the training set and 0.842 in the test set, which was better than that of ModelC (training set: P＜0.001; test set: P=0.032) and ModelR (training set: P＜0.001; test set: P=0.044). The AUC of ModelCom in the training set and test set were 0.951 and 0.881, respectively, which was better than that of ModelD in the training set (P=0.012), but there was no significant difference in the test set (P=0.157). The calibration curve showed that ModelCom had a good goodness of fit (Hosmer-Lemeshow test, P=0.402 for training set, P=0.689 for test set). Decision curve analysis showed that the net benefit of ModelCom in identifying positive MVI and negative  MVI was higher than that of other models. 
Conclusion The ModelCom based on CNN can accurately predict the MVI status of HCC.


Key words: liver neoplasms, convolutional neural network, radiomics

版权所有 © 《河北医科大学学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发