基于多序列MRI影像组学预测早期宫颈癌淋巴血管侵犯的研究

王海波; 崔薇; 杨玮丽

doi:10.16766/j.cnki.issn.1674-4152.002244

基于多序列MRI影像组学预测早期宫颈癌淋巴血管侵犯的研究

doi: 10.16766/j.cnki.issn.1674-4152.002244

1.
宁波大学附属人民医院放射科, 浙江宁波 315100
2.
宁波大学附属人民医院妇产科

基金项目:

浙江省医药卫生科技计划项目 2019RC274

详细信息

通讯作者:
王海波，E-mail: wang800080@163.com

中图分类号: R737.33R445.2
计量
- 文章访问数: 292
- HTML全文浏览量: 85
- PDF下载量: 5
- 被引次数: 0
出版历程
- 收稿日期: 2021-04-13
- 网络出版日期: 2022-03-02

Multi-sequence MRI-based radiomics predicting lymph-vascular space invasion in early-stage cervical cancer

1.
Department of Radiology, the Affiliated People's Hospital of Ningbo University, Ningbo, Zhejiang 315100, China

摘要

摘要: 目的探索基于多序列MRI图像的影像组学方法在预测早期宫颈癌淋巴血管侵犯(LVSI)中的临床价值。方法选取2015年1月—2020年2月宁波大学附属人民医院收治经病理证实的早期宫颈癌患者134例，分层抽样选取训练组91例，验证组43例，术前均行MRI平扫、对比增强(CE-MRI)及弥散成像(DWI)检查。在T₂WI-FS、CE-MRI及DWI序列图像上分别勾画肿瘤感兴趣区，采用LASSO回归及诺模图法提取影像组学特征并建立预测模型，训练组进行特征选择分类及模型建立，验证组对构建的预测模型进行验证，分析基于MRI各序列影像组学模型对早期宫颈癌LVSI的预测效能。结果采用LASSO回归在早期宫颈癌患者的T₂WI-FS、CE-MRI和DWI序列影像中分别提取具有预测意义的宫颈癌LVSI的影像组学特征，WavEnLH_s-4、Horzl_LngREmph在各序列中均被筛选出。通过logistics回归模型分别构建的不同序列MRI的影像组学模型对早期宫颈癌LVSI诊断效能均较高，T₂WI-FS、CE-MRI及DWI在训练组的AUC分别为0.810、0.803和0.781，在验证组的AUC分别为0.785、0.761和0.752。使用诺模图法构建的多序列MRI影像组学在训练组的AUC为0.893，在验证组的AUC为0.859。结论通过诺模图法构建的基于T₂WI-FS、CE-MRI及DWI序列影像组学模型作为一种客观的影像分析方法，对早期宫颈癌LVSI具有较高的预测效能并具有一定临床应用价值。
- 影像组学 /
- 磁共振成像 /
- 淋巴血管侵犯 /
- 早期宫颈癌
Abstract: Objective To explore the clinical value of multi-sequence MRI-based radiomics in predicting lymph-vascular space invasion (LVSI) in early-stage cervical cancer. Methods A total of 134 patients (91 cases in the training group and 43 cases in the validation group) with pathological early-stage cervical cancer in the Affiliated People's Hospital of Ningbo University from January 2015 to February 2020, were retrospectively collected. All patients underwent MRI plain scan, contrast-enhanced MRI (CE-MRI) and diffusion-weighted imaging (DWI) before surgery. MRI images of each sequence were obtained, and the region of interest was drawn. Radiomic features were extracted by using the least absolute shrinkage and selection operator (LASSO) and nomogram method to construct the predictive model. The training group was used to extract feature, establish signature and construct the predictive model. The validation group was used to verify the predictive model. Receiver operating characteristic curve was used to analyse the predictive effect of each sequence MRI-based radiomic model on LVSI in early-stage cervical cancer. Results The predictive LVSI features were extracted from T₂WI-FS, CE-MRI and DWI sequence images in patients with early-stage cervical cancer by LASSO regression. WavEnLH_s-4 and Horzl_LngREmph were all screened out. Results showed that the diagnostic efficiency of multi-sequence MRI imaging models constructed by the Logistic regression model was high for LVSI in early-stage cervical cancer. AUC of T₂WI-FS, CE-MRI and DWI was 0.810, 0.803 and 0.781, respectively, in the training group and 0.785, 0.761 and 0.752, respectively, in the verification group. The AUC of multi-sequence MRI-based radiomic model constructed by nomogram was 0.893 in the training group and 0.859 in the verification group. Conclusion As an objective image analysis method, the nomographic model based on T₂WI-FS, CE-MRI and DWI sequence has a high predictive effect and certain clinical application value in LVSI of early-stage cervical cancer.
- Radiomics /
- Magnetic resonance imaging /
- Lymph-vascular space invasion /
- Early-stage cervical cancer

HTML全文

图 1 宫颈癌ROI选取示意图

注：A~C分别示轴位T₂WI-FS、矢状位T₁WI增强及轴位DWI序列图像中宫颈癌病变区域的ROI选取。

下载: 全尺寸图片幻灯片

图 2 宫颈癌ROI影像组学特征参数提取

注：A~C分别示于轴位T₂WI-FS、矢状位T₁WI增强及轴位DWI序列图像对宫颈癌ROI所提取的影像组学特征参数示意。

下载: 全尺寸图片幻灯片

表 1 2组早期宫颈癌患者临床特征比较

组别	例数	年龄(x±s, 岁)	FIGO分期[例(%)]			病理类型[例(%)]					LVSI[例(%)]
组别	例数	年龄(x±s, 岁)	Ⅰa	Ⅰb	Ⅱa	腺癌	鳞癌	腺鳞癌	小细胞癌	其他癌	有	无
训练组	91	52.6±14.2	25(27.5)	34(37.4)	32(35.1)	44(48.3)	24(26.4)	13(14.3)	5(5.5)	5(5.5)	53(58.2)	38(41.7)
验证组	43	50.3±12.3	10(23.2)	18(41.9)	15(34.9)	19(44.2)	12(27.9)	5(11.6)	4(9.3)	3(7.0)	26(60.5)	17(39.5)
统计量		0.926^a	0.352^b			1.277^b					0.060^b
P值		0.356	0.839			0.888					0.852
注：^a为t值，^b为χ²值。

下载: 导出CSV

表 2 T₂WI-FS、CE-MRI及DWI序列LASSO降维后提取的纹理特征参数

检查方法	纹理特征参数
T₂WI-FS	S(1, 0)SumOfSqs；S(3, -3)SumEntrp；S(1, -1)DifVarnc；WavEnLH_s-4；Vertl_RLNonUni；Teta1；135dr_GLevNonU；45dgr_RLNonUni；Horzl_LngREmph；S(4, -4)DifVarnc
CE-MRI	Teta3；S(5, 0)InvDfMom；45dgr_RLNonUni；WavEnLL_s-4；S(4, 4)Entropy；Horzl_LngREmph；S(0, 3)DifVarnc；S(2, -2)SumVarnc；S(2, 0)InvDfMom；WavEnHL_s-5
DWI	Teta4；S(3, -3)Entropy；WavEnLL_s-4；S(2, 2)DifEntrp；S(1, -1)DifVarnc；S(0, 1)SumOfSqs；135dr_RLNonUni；Sigma；Vertl_RLNonUni；Horzl_LngREmph

下载: 导出CSV

表 3 各序列MRI影像组学预测宫颈癌LVSI的效能

组别	T₂WI-FS				CE-MRI
组别	AUC(95% CI)	准确率(%)	灵敏度(%)	特异度(%)	AUC(95% CI)	准确率(%)	灵敏度(%)	特异度(%)
训练组	0.810(0.667~0.894)	82.8	83.3	80.9	0.803(0.716~0.899)	82.3	82.7	81.7
验证组	0.785(0.623~0.870)	79.1	81.4	77.4	0.761(0.643~0.852)	75.1	78.7	76.5

组别	DWI				多序列诺模图
组别	AUC(95% CI)	准确率(%)	灵敏度(%)	特异度(%)	AUC(95% CI)	准确率(%)	灵敏度(%)	特异度(%)
训练组	0.781(0.671~0.880)	78.2	77.6	73.8	0.893(0.799~0.997)	91.1	91.8	88.9
验证组	0.752(0.631~0.832)	72.0	74.5	70.7	0.859(0.741~0.930)	86.2	87.8	85.1

下载: 导出CSV

参考文献(17)

[1]	BRAY F, FERLAY J, SOERJOMATARAM I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA: Cancer J Clin, 2018, 68(6): 394-424. doi: 10.3322/caac.21492
[2]	SHIM S H, LIM M C, KIM H J, et al. Can simple trachelectomy or conization show comparable survival rate compared with radical trachelectomy in IA1 cervical cancer patients with lymphovascular space invasion who wish to save fertility? A systematic review and guideline recommendation[J]. PLoS One, 2018, 13(1): e0189847. doi: 10.1371/journal.pone.0189847
[3]	LI M, WU S W, XIE Y Q, et al. Cervical invasion lymphovascular space invasion, and ovarian metastasis as predictors of lymph node metastasis and poor outcome on stages Ⅰ to Ⅲ endometrial cancers: A single-center retrospective study[J]. World J Surg Oncol, 2019, 17(1): 193. doi: 10.1186/s12957-019-1733-2
[4]	PATEL-LIPPMANN K, ROBBINS J B, BARROIHET L, et al. MR imaging of cervical cancer[J]. Magn Reson Imaging Clin N Am, 2017, 25(3): 635-649. doi: 10.1016/j.mric.2017.03.007
[5]	XIAO M, YAN B, LI Y, et al. Diagnostic performance of MR imaging in evaluating prognostic factors in patients with cervical cancer: A meta-analysis[J]. Eur Radiol, 2020, 30(3): 1405-1418. doi: 10.1007/s00330-019-06461-9
[6]	XIAO M, MA F, LI Y, et al. Multiparametric MRI-based radiomics nomogram for predicting lymph node metastasis in early-stage cervical cancer[J]. J Magn Reson Imaging, 2020, 52(3): 885-896. doi: 10.1002/jmri.27101
[7]	FANG J, ZHANG B, WANG S, et al. Association of MRI-derived radiomic biomarker with disease-free survival in patients with early-stage cervical cancer[J]. Theranostics, 2020, 10(5): 2284-2292. doi: 10.7150/thno.37429
[8]	LÓPEZ-CARBALLEIRA A, BALEATO-GONZÁLEZ S, GARCIA-FIGUEIRAS R, et al. Magnetic resonance imaging for staging and treatment planning in cervical cancer[J]. Radiologia, 2016, 58(4): 268-276. doi: 10.1016/j.rx.2015.11.008
[9]	NOMDEN C N, PÖTTER R, D E LEEUW A A C, et al. Nodal failure after chemo-radiation and MRI guided brachytherapy in cervical cancer: Patterns of failure in the EMBRACE study cohort[J]. Radiother Oncol, 2019, 134: 185-190. doi: 10.1016/j.radonc.2019.02.007
[10]	霍文礼, 由英宁, 卢再鸣, 等. 18F-FDG PET/CT纹理特征联合MMP-7表达预测早期宫颈癌盆腔淋巴结转移[J]. 临床放射学, 2021, 40(1): 170-174. https://www.cnki.com.cn/Article/CJFDTOTAL-LCFS202101039.htm
[11]	LU Y, PENG W, SONG J, et al. On the potential use of dynamic contrast-enhanced (DCE) MRI parameters as radiomic features of cervical cancer[J]. Med Phys, 2019, 46(11): 5098-5109. doi: 10.1002/mp.13821
[12]	JALAGUIER-COUDRAY A, VILLARD-MAHJOUB R, DELOUCHE A, et al. Value of dynamic contrast enhanced and diffusion weighted MR imaging in the detection of pathologic complete response in cervical cancer after neoadjuvant therapy: A retrospective observational study[J]. Radiology, 2017, 284(2): 432-442. doi: 10.1148/radiol.2017161299
[13]	PENG J, WANG W, ZENG D. Application of magnetic resonance imaging in diagnosis of Uterus Cervical Carcinoma[J]. J Xray Sci Technol, 2017, 25(2): 205-211.
[14]	LI Z, LI H, WANG S, et al. MR-Based radiomics nomogram of cervical cancer in prediction of the lymph-vascular space invasion preoperatively[J]. J Magn Reson Imaging, 2019, 49(5): 1420-1426. doi: 10.1002/jmri.26531
[15]	WANG T, GAO T, GUO H, et al. Preoperative prediction of parametrial invasion in early-stage cervical cancer with MRI-based radiomics nomogram[J]. Eur Radiol, 2020, 30(6): 3585-3593. doi: 10.1007/s00330-019-06655-1
[16]	LIN Y C, LIN C H, LU H Y, et al. Deep learning for fully automated tumor segmentation and extraction of magnetic resonance radiomics features in cervical cancer[J]. Eur Radiol, 2020, 30(3): 1297-1305. doi: 10.1007/s00330-019-06467-3
[17]	VARGHESE B A, CHEN F, HWANG D H, et al. Differentiation of predominantly solid enhancing lipid-poor renal cell masses by use of contrast-enhanced CT: Evaluating the role of texture in tumor subtyping[J]. AJR Am J Roentgenol, 2018, 211(6): W288-296. https://pubmed.ncbi.nlm.nih.gov/30240299/