Please wait a minute...
浙江大学学报（医学版）
浙江大学学报（医学版）  2017, Vol. 46 Issue (5): 492-497    DOI: 10.3785/j.issn.1008-9292.2017.10.07
精准影像医学专题     
磁共振成像强化信号特征预测胶质母细胞瘤EGFR基因扩增状态的影像组学研究
董飞1, 李倩1, 蒋飚1, 曾强2, 华建明1, 张敏鸣1
1. 浙江大学医学院附属第二医院放射科, 浙江 杭州 310009;
2. 浙江大学医学院附属第二医院神经外科, 浙江 杭州 310009
Quantitative analysis of enhanced MRI features for predicting epidermal growth factor receptor gene amplification in glioblastoma multiforme with radiomic method
DONG Fei1, LI Qian1, JIANG Biao1, ZENG Qiang2, HUA Jianming1, ZHANG Minming1
1. Department of Radiology, the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, China;
2. Department of Neurosurgery, the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, China
 全文: PDF(1035 KB)  
摘要：

目的：探讨采用影像组学研究方法分析胶质母细胞瘤磁共振强化信号特征预测表皮生长因子受体（EGFR）基因扩增状态的可行性。方法：收集术前行颅脑平扫及增强磁共振检查的80例经常规病理学检查确诊为胶质母细胞瘤并行分子病理学EGFR基因扩增状态检测的患者，按3：2比例随机分组至训练数据集和测试数据集。利用半自动软件高通量提取患者增强磁共振图像中强化区及周围水肿区的定量信号特征，经主成分分析等数据处理后利用随机森林模型、支持向量机模型和神经网络模型进一步分析数据，以测试数据集模型的受试者工作特征曲线下面积（AUC）作为模型的评价指标。结果：共提取强化区及水肿区特征542个，经主成分分析得到48个主成分（累积贡献率为100%），选择其中31个主成分（累积贡献率为98.5%）建模。随机森林模型、支持向量机模型和神经网络模型在测试数据集的AUC值分别为0.74、0.69和0.63。结论：采用影像组学研究方法分析胶质母细胞瘤强化信号特征并选择合适的模型进行分析对胶质母细胞瘤EGFR基因扩增状态有一定的预测价值。
关键词： 胶质母细胞瘤/病理学;  受体;  基因扩增;  图像处理;  计算机辅助;  人工智能;  磁共振成像;  表皮生长因子    
Abstract:

Objective: To assess the value of contrast enhanced MRI features for predicting epidermal growth factor receptor (EGFR) gene amplification in glioblastoma multiforme (GBM) with radiomic method.Methods: Eighty patients with EGFR status examined GBM were retrospectively reviewed. The data were randomly divided into a training dataset (60%) and test dataset (40%). Texture features of each case were extracted from the enhanced region and the edema region in contrast enhanced MR images. Principal component analysis was used for dimension reduction. Random forest model, support vector machine model and neural network model were built. Area under the curve (AUC) of the receiver operating characteristics curve was used to assess the performance of models with test dataset.Results: A total of 542 features were extracted from the enhanced region and the edema region. Forty-eight principal components were obtained, which accounted for 100% accumulation contribution rate, and the first 31 principal components were selected for models building, which accounted for 98.5% accumulation contribution rate. The values of AUCs were 0.74, 0.69 and 0.63 for random forest model, support vector machine model and neural network model in the test dataset, respectively.Conclusion: Radiomic method with proper model may have a potential role in predicting the EGFR gene status with enhanced MRI features derived from the enhanced region and the edema region in patients with glioblastoma multiforme.
Key words: Gene amplification    Magnetic resonance imaging    Artificial intelligence    Receptor,epidermal growth factor    Glioblastoma/pathology    Image processing,computer-assisted
收稿日期: 2017-05-08 出版日期: 2017-10-25
CLC:  R445.2  
基金资助:

浙江省医药卫生科技计划（2016KYA104，2017KY374，2017205359）；国家重点研发计划（2016YFC13066）；国家卫生和计划生育委员会科研基金（2016149022）；国家自然科学基金（8157165）
通讯作者: 张敏鸣(1957-),女,博士,教授,主任医师,博士生导师,主要从事神经退行性疾病的多模态影像和肿瘤精准影像学研究;E-mail:zhangminming@zju.edu.cn;http://orcid.org/0000-0003-0145-7558     E-mail: zhangminming@zju.edu.cn
作者简介: 董飞(1983-),男,硕士,主治医师,主要从事放射诊断学研究;E-mail:dngfei2@163.com;http://orcid.org/0000-0001-9533-385X
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

董飞 等. 磁共振成像强化信号特征预测胶质母细胞瘤EGFR基因扩增状态的影像组学研究[J]. 浙江大学学报（医学版）, 2017, 46(5): 492-497.

DONG Fei, LI Qian, JIANG Biao, ZENG Qiang, HUA Jianming, ZHANG Minming. Quantitative analysis of enhanced MRI features for predicting epidermal growth factor receptor gene amplification in glioblastoma multiforme with radiomic method. Journal of ZheJiang University(Medical Science), 2017, 46(5): 492-497.

链接本文:

http://www.zjujournals.com/xueshu/med/CN/10.3785/j.issn.1008-9292.2017.10.07    或    http://www.zjujournals.com/xueshu/med/CN/Y2017/V46/I5/492

[1] WIRSCHING H G, GALANIS E, WELLER M. Glioblastoma[J]. Handb Clin Neurol,2016,134:381-397.
[2] ALDAPE K, ZADEH G, MANSOURI S, et al. Glioblastoma:pathology, molecular mechanisms and markers[J]. Acta Neuropathol,2015,129(6):829-848.
[3] HALATSCH M E, SCHMIDT U, BEHNKE-MURSCH J, et al. Epidermal growth factor receptor inhibition for the treatment of glioblastoma multiforme and other malignant brain tumours[J]. Cancer Treat Rev,2006,32(2):74-89.
[4] RAMIS G, THOMÀS-MOYÀ E, DE MATTOS S F, et al. EGFR inhibition in glioma cells modulates Rho signaling to inhibit cell motility and invasion and cooperates with temozolomide to reduce cell growth[J/OL]. PLoS One,2012,7(6):e38770.
[5] BRENNAN C W, VERHAAK R G, MCKENNA A, et al. The somatic genomic landscape of glioblastoma[J]. Cell,2013,155(2):462-477.
[6] LIU F, HON G C, VILLA G R, et al. EGFR mutation promotes glioblastoma through epigenome and transcription factor network remodeling[J]. Mol Cell,2015,60(2):307-318.
[7] VERHAAK R G, HOADLEY K A, PURDOM E, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1[J]. Cancer Cell,2010,17(1):98-110.
[8] YOUNG R J, GUPTA A, SHAH A D, et al. Potential role of preoperative conventional MRI including diffusion measurements in assessing epidermal growth factor receptor gene amplification status in patients with glioblastoma[J]. AJNR Am J Neuroradiol,2013,34(12):2271-2277.
[9] GUTMAN D A, DUNN W D, GROSSMANN P, et al. Somatic mutations associated with MRI-derived volumetric features in glioblastoma[J]. Neuroradiology,2015,57(12):1227-1237.
[10] LAL A, GLAZER C A, MARTINSON H M, et al. Mutant epidermal growth factor receptor up-regulates molecular effectors of tumor invasion[J]. Cancer Res,2002,62(12):3335-3339.
[11] GOLDMAN C K, KIM J, WONG W L, et al. Epidermal growth factor stimulates vascular endothelial growth factor production by human malignant glioma cells:a model of glioblastoma multiforme pathophysiology[J]. Mol Biol Cell,1993,4(1):121-133.
[12] MABRAY M C, BARAJAS R F, CHA S. Modern brain tumor imaging[J]. Brain Tumor Res Treat,2015,3(1):8-23.
[13] HUANG Y Q, LIANG C H, HE L, et al. Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer[J]. J Clin Oncol,2016,34(18):2157-2164.
[14] LARUE R T, DEFRAENE G, DE RUYSSCHER D, et al. Quantitative radiomics studies for tissue characterization:a review of technology and methodological procedures[J]. Br J Radiol,2017,90(1070):20160665.
[1] 许晶晶 等. 影像学在肿瘤精准医疗时代的机遇和挑战[J]. 浙江大学学报（医学版）, 2017, 46(5): 455-461.
[2] 杨荣 等. Molday IONTM EverGreen标记大鼠骨髓内皮祖细胞及体外磁共振成像研究[J]. 浙江大学学报（医学版）, 2017, 46(5): 481-486.
[3] 张思影 等. CT和磁共振参数反应图在肿瘤精准疗效评估中的研究进展[J]. 浙江大学学报（医学版）, 2017, 46(5): 468-472.
[4] 潘瑶 等. 胰腺癌的影像学精准诊断与评估[J]. 浙江大学学报（医学版）, 2017, 46(5): 462-467.
[5] 李爱静 等. 动态增强磁共振成像参照物模型定量参数与乳腺癌预后因素及分子病理分型的关系[J]. 浙江大学学报（医学版）, 2017, 46(5): 505-510.
[6] 裴磊 等. 磁共振平扫T1加权像脑内核团高信号与钆对比剂注射次数的相关性[J]. 浙江大学学报（医学版）, 2017, 46(5): 487-491.
[7] 王苏波 等. 动态对比增强磁共振药代动力学模型在鉴别富细胞型子宫平滑肌瘤中的应用[J]. 浙江大学学报（医学版）, 2017, 46(5): 498-504.
[8] 王海凤 等. CXC趋化因子受体4通过S期激酶相关蛋白2调控乳腺癌细胞周期的机制[J]. 浙江大学学报（医学版）, 2017, 46(4): 357-363.
[9] 陈志强 等. 不同量甲醛固定液对荧光原位杂交法检测乳腺原发性浸润癌HER2基因扩增的影响[J]. 浙江大学学报（医学版）, 2017, 46(4): 439-444.
[10] 方兵 等. 舌下神经管区硬脑膜动静脉瘘介入治疗二例[J]. 浙江大学学报（医学版）, 2017, 46(4): 445-448.
[11] 温弘 等. 产前诊断Joubert综合征一例并文献复习[J]. 浙江大学学报（医学版）, 2017, 46(3): 274-278.
[12] 何玉洁,潘建平. 病原菌对NOD样受体及Toll样受体信号通路介导的固有免疫逃逸机制研究进展[J]. 浙江大学学报（医学版）, 2017, 46(2): 218-224.
[13] 何玉洁 等. 病原菌对NOD样受体及Toll样受体信号通路介导的固有免疫逃逸机制研究进展[J]. 浙江大学学报（医学版）, 2017, 46(2): 218-224.
[14] 王颖,汪仪,陈忠. 中枢胆碱能系统与癫痫关系的研究进展[J]. 浙江大学学报（医学版）, 2017, 46(1): 15-21.
[15] 吴菡,王钟瑾,明文杰,王爽,丁美萍. 长程视频脑电图监测癫痫患者发作间期痫样放电的时段分析[J]. 浙江大学学报（医学版）, 2017, 46(1): 30-35.