Page 112 - 磁共振成像2024年7期电子刊

磁共振成像 2024年7月第15卷第7期 Chin J Magn Reson Imaging, Jul, 2024, Vol. 15, No. 7 临床研究｜｜Clinical Articles

IVIM、DKI联合DCE-MRI的影像组学在预测

乳腺癌HER-2表达状态中的应用价值

1
1
3
赵晓萌 ，邵硕 ，郑宁 ，崔景景 ，刘诗晗 ，吴建伟 1
2*
2
作者单位 1.济宁医学院临床医学院，济宁 272013；2.济宁市第一人民医院磁共振室，济宁 272000；3.联影智能医疗科技（北京）有
限公司，北京 100089
* 通信作者 郑宁，E-mail: zhengning_369@163.com
中图分类号 R445.2；R737.9 文献标识码 A DOI 10.12015/issn.1674-8034.2024.07.018
本文引用格式 赵晓萌, 邵硕, 郑宁, 等. IVIM、DKI联合DCE-MRI的影像组学在预测乳腺癌HER-2表达状态中的应用价值[J]. 磁
共振成像, 2024, 15(7): 105-111.
[摘要] 目的 探讨联合体素内不相干运动（intravoxel incoherent motion, IVIM）、扩散峰度成像（diffusion kurtosis imaging,
DKI）和动态对比增强磁共振成像（dynamic contrast-enhanced magnetic resonance imaging, DCE-MRI)、参数图构建影像组学模
型预测乳腺癌患者人类表皮生长因子受体 2（human epidermal growth factor receptor 2, HER-2）的表达状态。材料与方法 回顾
性分析 192例乳腺癌患者病例资料，根据患者的病理结果分为 HER-2表达阳性组（48例）和 HER-2表达阴性组（144例），术前
均行 IVIM、DKI及 DCE-MRI。并按照 8∶2 的比例将病例随机分为训练集（154 例）和测试集（38 例）。在灌注分数（perfusion
fraction, f）、灌注相关扩散系数（perfusion related diffusion coefficient, D ）、真实扩散系数（real diffusion coefficient, D）、平均扩
*
散率（mean diffusivity, MD）和平均扩散峰度值（mean kurtosis, MK）参数图和第 2期 DCE-MRI（DCE-2）图像中勾画出病变区
域的三维感兴趣区（region of interest, ROI），并提取其中的影像组学特征。采用 Z分数归一化对特征进行标准化处理，并使用 K
最佳、最小冗余最大相关（max-relevance and min-redundancy, mRMR）、最小绝对收缩与选择算子回归（least absolute shrinkage
and selection operator, LASSO）算法依次对特征进行降维和选择，通过 logistic 逻辑回归（logistic regression, LR）分类器分别建
立参数图模型及联合模型，并采用 5 折交叉验证法验证模型的稳定性。通过受试者工作特征 （receive operating characteristic,
ROC）曲线和曲线下面积（area under the curve, AUC）对不同参数图像模型及联合模型的诊断效能进行分析，使用 DeLong检验
对各模型间 ROC曲线进行比较，使用决策曲线分析（decision curve analysis, DCA）对模型的临床价值进行评估。结果 从每个
*
ROI中提取了 2286个 MRI特征，在 f、D 、D、MD、MK 参数图、第 2期 DCE-MRI和联合序列中分别筛选得到 7、6、7、6、7、
12、10 个特征与 HER-2 表达状态相关。f、D 、D、MD、MK 参数图模型及第 2 期 DCE 模型在测试集中的 AUC 分别为 0.693、
*
0.679、0.586、0.682、0.661、0.732；联合模型在测试集中的 AUC 为 0.861（95% CI：0.775~0.958），敏感度和特异度分别为
*
100.0% 和 71.4%，经 DeLong 检验，训练集中联合模型与 f、D、D 、MD、MK 参数图模型及 DCE-2 模型之间 AUC 差异均有统
计学意义（P均<0.05）。结果表明联合模型对预测HER-2的表达状态优于单一模型。结论 基于DCE-MRI、IVIM和DKI的影像
组学联合模型可以在术前有效预测乳腺癌患者的 HER-2 表达状态，有助于临床对乳腺癌进行诊断、分型、制订治疗方案及
预后。
[关键词] 人类表皮生长因子受体2；乳腺癌；扩散峰度成像；体素内不相干运动；影像组学；磁共振成像
Application value of IVIM, DKI and DCE-MRI radiomics predicting HER-2 expression in breast
cancer

2
1
1
3
2*
ZHAO Xiaomeng , SHAO Shuo , ZHENG Ning , CUI Jingjing , LIU Shihan , WU Jianwei 1
2
1 Clinical Medical College, Jining Medical University, Jining 272013, China; Magnetic Resonance Imaging Room, Jining First People's
3
Hospital, Jining 272000, China; United Imaging Intelligence Medical Technology Co., Ltd., Beijing 100089, China
* Correspondence to ZHENG N, E-mail: zhengning_369@163.com
Received 19 Mar 2024, Accepted 6 Jun 2024; DOI 10.12015/issn.1674-8034.2024.07.018
ACKNOWLEDGMENTS Key Research and Development Program of Jining (No. 2023YXNS117).
Cite this article as ZHAO X M, SHAO S, ZHENG N, et al. Application value of IVIM, DKI and DCE-MRI radiomics predicting
HER-2 expression in breast cancer[J]. Chin J Magn Reson Imaging, 2024, 15(7): 105-111.
Abstract Objective: To explore the intravoxel incoherent motion (IVIM), diffusion kurtosis imaging (DKI) and diagnostic value of
radiomics models based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), in prediction of human epidermal growth
factor receptor 2 (HER-2) positive status in breast cancer patients. Materials and Methods: The clinical data of 192 patients with breast cancer
were analyzed retrospectively. Patients were divided into HER-2 positive group (48 cases) and HER-2 negative group (144 cases) based on
their pathological results. All patients underwent IVIM, DKI, and DCE-MRI scans before surgery. And then these data were randomly divided
into training sets (n=154) and test sets (n=38) at a ratio of 8∶2. The three-dimensional volume region of interest of the tumor was manually
*
delineated on the perfusion fraction (f), perfusion related diffusion coefficient (D ), real diffusion coefficient (D), mean diffusivity (MD) and
mean kurtosis (MK) parameter maps and the second phase of dynamic contrast-enhanced MRI, and radiomics features were extracted. The
Z-score normalization was used for feature normalization, and the Select K Best, max-relevance and min-redundancy (mRMR) and least
absolute shrinkage and selection operator (LASSO) were used to single out the most valuable radiomic features. The parametric map models
and a combined model were established by logistic regression (LR) classifier, and the stability of the models was verified by the 5-fold
cross-validation. The receive operating characteristic (ROC) curve and area under the curve (AUC) were used to evaluate the efficacy of the
model. In addition, the DeLong test was used to compare the models, and decision curve analysis (DCA) was used to evaluate the models.
收稿日期 2024-03-19 接受日期 2024-06-06
基金项目 济宁市重点研发计划项目（编号：2023YXNS117）

https://www.chinesemri.com ·105 ·