[1]杨日辉,范伟雄,董婷,等.多序列MR影像组学融合模型预测胶质瘤分级的价值[J].中国医学物理学杂志,2023,40(5):545-549.[doi:DOI:10.3969/j.issn.1005-202X.2023.05.003]
　YANG Rihui,FAN Weixiong,DONG Ting,et al.Predictive value of multi-sequence MRI radiomics fusion model for glioma grading[J].Chinese Journal of Medical Physics,2023,40(5):545-549.[doi:DOI:10.3969/j.issn.1005-202X.2023.05.003]
点击复制

多序列MR影像组学融合模型预测胶质瘤分级的价值()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
40卷
期数:
2023年第5期
页码:
545-549
栏目:
医学影像物理
出版日期:
2023-05-26

文章信息/Info

Title:
Predictive value of multi-sequence MRI radiomics fusion model for glioma grading
文章编号:
1005-202X（2023）05-0540-05
作者:
杨日辉1; 2; 范伟雄2; 董婷3; 江桂华1; 3
1.广东医科大学， 广东 湛江 524023； 2.梅州市人民医院（梅州市医学科学院）， 广东 梅州 514031； 3.广东省第二人民医院影像科， 广东 广州 510317
Author(s):
YANG Rihui1; 2;  FAN Weixiong2;  DONG Ting3;  JIANG Guihua1;  3
1. Guangdong Medical University, Zhanjiang 524023, China 2. Meizhou Peoples Hospital (Meizhou Academy of Medical Sciences), Meizhou 514031, China 3. Department of Imaging, Guangdong Second Provincial General Hospital, Guangzhou 510317, China
关键词:
胶质瘤; 病理分级; 影像组学; 融合模型
Keywords:
glioma pathological grade radiomics fusion model
分类号:
R742；R816.1
DOI:
DOI:10.3969/j.issn.1005-202X.2023.05.003
文献标志码:
A
摘要:
目的：探讨多序列MR影像组学特征及融合模型预测胶质瘤分级的价值。方法：回顾性分析2016年1月~2021年6月间梅州市人民医院经病理证实且行MR检查的胶质瘤病人资料。将多序列MR DICOM格式的原始图像导入ITK-SNAP软件进行VOI勾画，采用GE A.K分析软件进行影像组学特征提取，用ANOVA+Mann-Whiney、Spearman相关分析和LASSO模型进行特征筛选。选择Logistic回归（LR）算法构建单序列模型，选择LR、线性判别分析（LDA）、支持向量机（SVM）3种机器学习算法构建融合序列模型。通过受试者操作特征曲线评估不同模型的预测效能，并计算其曲线下面积（AUC）。结果：共有150例患者纳入本研究，其中低级别胶质瘤（LGG）41例，高级别胶质瘤（HGG）组109例。LGG组与HGG组病人间的性别、年龄差异无统计学意义（P>0.05）。共获得5个最优特征集，其中T1WI特征集包含15个特征，T2WI特征集包含31个特征，Flair特征集包含25个特征，DWI特征集包含12个特征，T1C包含4个特征，融合序列集包含44个特征。基于T1WI、T2WI、Flair、DWI及T1C各序列MR影像组学预测模型在训练集和测试集中的AUC分别为0.719 0、0.651 4，0.769 5、0.711 4，0.7410、0.610 2，0.721 9、0.747 0，0.815 7、0.754 5。基于LR、LDA、SVM算法的融合模型在训练集和测试集中的AUC分别为0.952 4、0.767 0，0.894 8、0.688 1，0.928 6、0.704 5。结论：基于多序列MRI影像组学特征的单序列模型中，T1C单序列预测模型对胶质瘤分级诊断效能最高；相较于单序列预测模型，多序列融合预测模型具有更高的诊断效能，且基于LR算法构建的融合模型较LDA算法及SVM算法构建的融合模型显示出更高的预测效能。
Abstract:
glioma grading. Methods The data of glioma patients who were pathologically confirmed and underwent MR examination in Meizhou Peoples Hospital from January 2016 to June 2021 were retrospectively analyzed. The original multi-sequence MR images in DICOM format were imported into ITK-SNAP software for VOI delineation. After extracting radiomics signatures with GE A.K analysis software, ANOVA+Mann Whitney, Spearman correlation analysis and LASSO model were used for feature screening. Logistic regression (LR) algorithm was selected to build a single sequence model, while LR, LDA and SVM were adopted to establish fusion sequence models. The prediction performances of different models were evaluated using the receiver operating characteristic curve, and the area under the curve (AUC) was calculated. Results A total of 150 patients were enrolled in the study, including 41 in low-grade glioma group and 109 in high-grade glioma group. There was no significant difference in gender and age between LGG group and HGG group (P>0.05). Five optimal feature sets were obtained, and there were 15, 31, 25, 12, 4 and 44 features in T1WI feature set, T2WI feature set, Flair feature set, DWI feature set, T1C, and fusion sequence set. The AUC of MRI radiomics-based prediction models of T1WI, T2WI, Flair, DWI and T1C was 0.719 0, 0.769 5, 0.741 0, 0.721 9 and 0.815 7 on the training set, and 0.651 4, 0.711 4, 0.610 2, 0.747 0 and 0.754 5 on the test set, respectively. The AUC of the fusion models established with LR, LDA and SVM was 0.952 4, 0.894 8 and 0.928 6 on the training set, and 0.767 0, 0.688 1 and 0.704 5 on the test set, respectively. Conclusion Among the single sequence models based on multi-sequence MRI radiomics, T1C single sequence prediction model has the highest efficiency in glioma grading. Compared with single sequence prediction model, multi-sequence fusion prediction model has higher diagnostic efficiency, and the fusion model established with LR shows higher prediction efficiency than the fusion models established with LDA and SVM.

相似文献/References:

[1]谭泓叶,张晓红,张海黔.银纳米粒子对乏氧胶质瘤细胞的影响[J].中国医学物理学杂志,2017,34(1):89.[doi:10.3969/j.issn.1005-202X.2017.01.018]
　[J].Chinese Journal of Medical Physics,2017,34(5):89.[doi:10.3969/j.issn.1005-202X.2017.01.018]
[2]张国前,张书旭,王锐濠,等. Auto-planning在脑胶质瘤非共面容积调强放疗计划中的应用[J].中国医学物理学杂志,2018,35(5):514.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.004]
　ZHANG Guoqian,ZHANG Shuxu,WANG Ruihao,et al. Application research on Auto-planning in non-coplanar VMAT plan for brain gliomas[J].Chinese Journal of Medical Physics,2018,35(5):514.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.004]
[3]缪雨季,周倬,常青.甲氟喹对胶质瘤细胞的放射增敏作用[J].中国医学物理学杂志,2021,38(1):6.[doi:DOI:10.3969/j.issn.1005-202X.2021.01.002]
　MIAO Yuji,ZHOU Zhuo,CHANG Qing.The radiosensitization effect of mefloquine combined with X-ray on glioma cells[J].Chinese Journal of Medical Physics,2021,38(5):6.[doi:DOI:10.3969/j.issn.1005-202X.2021.01.002]
[4]邓娟,李昇霖,刘显旺,等.影像学评价大鼠胶质瘤及其微环境的研究进展[J].中国医学物理学杂志,2021,38(5):578.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.010]
　DENG Juan,,et al.Advances in research on imaging evaluations of rat glioma and its microenvironment[J].Chinese Journal of Medical Physics,2021,38(5):578.[doi:DOI:10.3969/j.issn.1005-202X.2021.05.010]
[5]张斌,薛彩强,林晓强,等.深度学习在脑胶质瘤影像学的研究进展[J].中国医学物理学杂志,2021,38(8):1048.[doi:DOI:10.3969/j.issn.1005-202X.2021.08.025]
　ZHANG Bin,XUE Caiqiang,LIN Xiaoqiang,et al.Advances in deep learning for brain glioma imaging[J].Chinese Journal of Medical Physics,2021,38(5):1048.[doi:DOI:10.3969/j.issn.1005-202X.2021.08.025]
[6]王瑞,齐崇,孟蓝熙,等.基于3D卷积网络和多模态MRI的脑胶质瘤自动分割[J].中国医学物理学杂志,2022,39(3):300.[doi:DOI:10.3969/j.issn.1005-202X.2022.03.007]
　WANG Rui,QI Chong,MENG Lanxi,et al.Auto-segmentation of glioma based on 3D convolutional network and multimodal MRI[J].Chinese Journal of Medical Physics,2022,39(5):300.[doi:DOI:10.3969/j.issn.1005-202X.2022.03.007]
[7]程秀,王俊,王瑞,等.APT、ASL及DCE-MRI在胶质瘤分级诊断中的应用[J].中国医学物理学杂志,2022,39(3):321.[doi:DOI:10.3969/j.issn.1005-202X.2022.03.011]
　CHENG Xiu,WANG Jun,et al.Application of APT, ASL and DCE-MRI in glioma grading[J].Chinese Journal of Medical Physics,2022,39(5):321.[doi:DOI:10.3969/j.issn.1005-202X.2022.03.011]

备注/Memo

备注/Memo:
【收稿日期】2022-12-23 【基金项目】国家自然科学基金（82271948） 【作者简介】杨日辉，研究方向：影像医学与核医学，E-mail: 15219165220@163.com 【通信作者】江桂华，博士，主任医师，博士生导师，E-mail: jianggh@gd2h.org.cn
更新日期/Last Update: 2023-05-26