多序列MRI影像组学对局部晚期宫颈鳞癌同步放化疗疗效的预测价值

doi:10.11958/20241488

天津医药 ›› 2025, Vol. 53 ›› Issue (2): 213-218.doi: 10.11958/20241488

多序列MRI影像组学对局部晚期宫颈鳞癌同步放化疗疗效的预测价值

田友军¹(), 谭正武², 杨柯¹, 彭剑敏¹, 陈红桃¹, 黄志平³^,^△()

1 天门市第一人民医院CT/MR室（邮编431700）
2 华中科技大学同济医学院附属协和医院放射科
3 天门市第一人民医院肿瘤科

收稿日期:2024-10-08 修回日期:2024-12-09 出版日期:2025-02-15 发布日期:2025-02-26
通讯作者: ^△ E-mail：806752023@qq.com
作者简介:田友军（1986），男，主治医师，主要从事放射诊断方面研究。E-mail：12619110328@qq.com

The predictive value of multi-sequence MRI radiomics in the therapeutic effect of concurrent chemoradiotherapy on locally advanced cervical squamous cell carcinoma

TIAN Youjun¹(), TAN Zhengwu², YANG Ke¹, PENG Jianmin¹, CHEN Hongtao¹, HUANG Zhiping³^,^△()

1 CT/MR Room, the First People's Hospital of Tianmen, Tianmen 431700, China
2 Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology
3 Department of Oncology, the First People's Hospital of Tianmen

Received:2024-10-08 Revised:2024-12-09 Published:2025-02-15 Online:2025-02-26
Contact: ^△ E-mail：806752023@qq.com

田友军, 谭正武, 杨柯, 彭剑敏, 陈红桃, 黄志平. 多序列MRI影像组学对局部晚期宫颈鳞癌同步放化疗疗效的预测价值[J]. 天津医药, 2025, 53(2): 213-218.

TIAN Youjun, TAN Zhengwu, YANG Ke, PENG Jianmin, CHEN Hongtao, HUANG Zhiping. The predictive value of multi-sequence MRI radiomics in the therapeutic effect of concurrent chemoradiotherapy on locally advanced cervical squamous cell carcinoma[J]. Tianjin Medical Journal, 2025, 53(2): 213-218.

摘要/Abstract

摘要：

目的观察多序列磁共振成像（MRI）影像组学对局部晚期宫颈鳞癌（CSCC）患者同步放化疗（CCRT）疗效的预测价值。方法选取行CCRT治疗的100例局部晚期CSCC患者的临床资料。按7∶3比例随机分为训练集（70例）与验证集（30例）。根据实体肿瘤疗效标准将患者分为完全缓解（CR）与部分缓解（PR）。收集所有患者治疗前横断面DWI、T₂WI及增强T₁WI延迟期的检查图像，使用ITK-SNAP软件包结合3个序列勾画感兴趣区（ROI），开源软件PyRadiomics提取影像组学特征。对MRI组学特征先采用最小冗余最大相关（mRMR）算法筛选出前30个主要特征后，采用基于10折交叉验证的最小绝对收缩和选择算子（Lasso）降维筛选非零系数特征，并根据训练集中Lasso-Logistic回归模型的加权系数计算患者组学标签；采用Logistic回归构建基于DWI、T₂WI及T₁WI各序列预测模型及多序列组学标签的预测模型；受试者工作特征（ROC）曲线评估各个组学模型对局部晚期CSCC患者CCRT疗效的预测价值。结果训练集CR组38例，PR组32例；验证集CR组16例，PR组14例。在训练集与验证集中，CR组与PR组患者的年龄、FIGO分期、分化程度、病灶最大径及月经情况差异均无统计学意义。从ROI靶区中共提取851个影像学特征，经mRMR算法保留前30个特征后，经Lasso-Logistic算法与10折交叉验证从每个单独序列各自的851个影像组学特征中筛选出3个与CR相关的特征。从3个序列联合后的2 553个特征中筛选出8个与CR相关的特征。ROC曲线结果显示，训练集与验证集中，多序列联合预测局部晚期CSCC患者CCRT治疗疗效的曲线下面积（AUC）分别为0.971、0.946，均高于T₁WI、T₂WI、DWI单序列预测（训练集：Z分别为2.683、2.046、2.817，P<0.05；验证集：Z分别为2.075、2.117、2.005，均P<0.05）。结论多序列MRI影像组学模型对局部晚期CSCC患者CCRT治疗疗效具有较高的预测价值。

关键词: 宫颈肿瘤, 癌, 鳞状细胞, 磁共振成像, 放化疗, 影像组学

Abstract:

Objective To observe the value of multi-sequence magnetic resonance imaging (MRI) radiomics in predicting the efficacy of concurrent chemoradiotherapy (CCRT) in locally advanced cervical squamous cell carcinoma (CSCC) patients. Methods Clinical data of 100 CSCC patients underwent CCRT treatment were selected. In order to better validate the performance of the model, patients were randomly divided into the training set (70 cases) and the validation set (30 cases) in a 7∶3 ratio. According to the efficacy criteria for solid tumors, patients were divided into the complete response (CR) group (n=16) and the partial response (PR) group (n=14). Examination images of cross-sectional DWI, T₂WI and enhanced T₁WI were collected from all patients before treatment. ITK-SNAP software package combined with three sequences were used to outline ROI, and the open source software PyRadiomics was used to extract image omics features. For MRI omics features, the minimum redundancy maximum correlation (mRMR) algorithm was used to analyze and screen out the first 30 main features, and then the minimum absolute contraction and selection method (Lasso) based on 10-fold cross-validation was used to reduce dimensionality to screen the non-zero coefficient features. According to the weighting coefficient of Lasso-Logistic regression model in the training set, patient omics labels were calculated. Logistic regression analysis was used to construct a prediction model based on DWI, T₂WI and T₁WI sequence prediction models and multiple sequenomics labels. Receiver operating characteristic (ROC) curves evaluated the predictive value of each omics model for CCRT treatment in patients with locally advanced CSCC. Results There were 38 cases in the CR group and 32 cases in the PR group in the training set. There were 16 cases in the CR group and 14 cases in the PR group in the validation set. There were no significant differences in patient age, FIGO stage, differentiation degree, maximum lesion diameter and menstrual status between the CR group and the PR group in the training and validation sets. A total of 851 imaging features were extracted from the ROI target area. After the first 30 features were retained by mRMR algorithm, 3 CR-related features were selected from the 851 imaging omics features of each individual sequence by Lasso algorithm and 10-fold cross-validation. Eight CR related features were selected from 2 553 features after the combination of the three sequences. ROC curve results showed that in the training set and validation set, the AUC of multiple sequences combined to predict the therapeutic effect of CCRT in patients with locally advanced CSCC was 0.971 and 0.946, respectively, which was higher than that of T₁WI, T₂WI and DWI single sequence prediction (training set Z=2.683, 2.046, 2.817, P<0.05; verification set Z=2.075, 2.117, 2.005, P<0.05). Conclusion The multi sequence MRI radiomics model has high predictive value for the efficacy of CCRT treatment in locally advanced CSCC patients.

Key words: uterine cervical neoplasms, carcinoma, squamous cell, magnetic resonance imaging, treatment outcome, radiomics

中图分类号:

R737.33

图/表 8

图1 局部晚期CSCC患者ROI勾画示意图 A、B、C分别为DWI、T2WI及T1WI序列图像下的ROI区域。

Fig.1 Diagrammatic drawing of ROI delineation for patients with locally advanced CSCC

表1 2组的训练集和验证集中的患者临床基线资料的比较

Tab.1 Comparison of clinical baseline data between the CR group and the PR group in the training set and the validation set

组别	训练集
组别	n		年龄/岁	FIGO分期（Ⅱ/Ⅲ/Ⅳ）	分化程度（低分化/中高分化）	病灶最大径/cm		月经情况（已绝经/未绝经）
CR组	38		53.66±4.26	10/17/11	11/27	4.52±0.65		30/8
PR组	32		53.63±5.05	10/15/7	9/23	4.33±0.53		25/7
t或χ²			0.030	0.503	0.006	1.323		0.007
组别	验证集
组别	n	年龄/岁		FIGO分期（Ⅱ/Ⅲ/Ⅳ）	分化程度（低分化/中高分化）		病灶最大径/cm	月经情况（已绝经/未绝经）
CR组	16	53.19±4.56		3/9/4	4/12		4.28±0.85	13/3
PR组	14	54.43±4.60		5/5/4	3/11		4.40±0.69	11/3
t或P		0.740		0.469^▲	1.000^▲		1.000	0.033^▲

表2 MRI不同序列影像组学特征及对应系数

Tab.2 Radiomics features of T1WI, T2WI and DWI and corresponding coefficients

MRI序列	特征分类	特征矩阵名称	特征名称	系数
DWI	original	glszm	ZoneEntropy	-0.005 0
	wavelet-LHL	glszm	SizeZoneNonUniformity	-0.425 0
	original	firstorder	90Percentile	-0.574 0
T₂WI	wavelet-LLL	glrlm	RunEntropy	-0.554 0
	original	shape	Maximum2DDiameterColumn	-0.347 6
	wavelet-LHL	glcm	MaximumProbability	-0.545 0
T₁WI	wavelet-HHL	glcm	JointEntropy	0.240 0
	original	firstorder	Range	-0.425 0
	wavelet-LHL	glszm	ZoneEntropy	-0.700 0
联合	original	shape	Flatness	0.135 2
	original	shape	MeshVolume	-0.133 6
	wavelet-HLL	firstorder	Minimum	0.086 6
	wavelet-LHL	glcm	Imc1	0.068 4
	wavelet-HLL	glcm	Imc2	-0.157 7
	wavelet-LHL	glcm	ClusterShade	0.297 1
	wavelet-HLL	ngtdm	Complexity	-0.117 3
	wavelet-HHL	glrlm	RunLengthNonUniformity	0.040 8

图2 训练集各序列影像组学评分分布图 A：T1WI；B：T2WI；C：DWI；D：多序列MRI。红色：PR患者；绿色：CR患者，X轴：患者数量，Y：评分分值。

Fig.2 Omics score distribution of each sequence image in the training set

图3 验证集各序列影像组学评分分布图 A：T1WI；B：T2WI；C：DWI；D：多序列MRI。红色：PR患者；绿色：CR患者，X轴：患者数量，Y：评分分值。

Fig.3 The distribution of omics scores of each sequence image in the validation set

表3 训练集中各序列影像组学模型对CCRT疗效的预测效能

Tab.3 Comparison of the predictive efficacy of radiomics model of each sequence in the training set for therapeutic effect of CCRT on patients with locally advanced CSCC

MRI序列	AUC	AUC 95%CI	敏感度/%	特异度/%
T₁WI	0.816	0.713~0.919	89.47	68.75
T₂WI	0.853	0.766~0.939	68.42	84.38
DWI	0.873	0.789~0.956	76.32	90.63
多序列联合	0.971	0.940~1.000	89.47	93.75

表4 验证集中各序列影像组学模型对CCRT疗效的预测效能

Tab.4 Comparison of the predictive efficacy of radiomics model of each sequence in the validation set for therapeutic effect of CCRT on patients with locally advanced CSCC

MRI序列	AUC	AUC 95%CI	敏感度/%	特异度/%
T₁WI	0.835	0.693~0.876	97.70	56.25
T₂WI	0.893	0.782~0.926	100.00	87.50
DWI	0.848	0.705~0.890	99.10	100.00
多序列联合	0.946	0.874~0.973	100.00	87.50

图4 各序列影像组学模型预测局部晚期CSCC患者CCRT疗效的ROC曲线 A:训练集；B：验证集。

Fig.4 ROC curves of radiomics model of each sequence in the training set and the validation set for therapeutic effect of CCRT on patients with locally advanced CSCC

参考文献 19

[1]	HE H, LIN C, LU Y, et al. Knockdown of miR-24 suppressed the tumor growth of cervical carcinoma through regulating PTEN/PI3K/AKT signaling pathway[J]. Biochem Genet, 2024, 62(2):1277-1290. doi:10.1007/s10528-023-10491-w.
[2]	王倩倩, 李婷芳, 王峰. CHD4调控端粒功能促进宫颈癌HeLa细胞生长的机制研究[J]. 天津医药, 2023, 51(9):909-915.
	WANG Q Q, LI T F, WANG F. Study on the mechanism of CHD4 regulating telomere function to promote cervical cancer HeLa cell proliferation[J]. Tianjin Med J, 2023, 51(9):909-915. doi:10.11958/20222091.
[3]	NASIOUDIS D, BYRNE M, KO E M, et al. Minimally invasive hysterectomy for stage IA cervical carcinoma:a survival analysis of the National Cancer Database[J]. Int J Gynecol Cancer, 2021, 31(8):1099-1103. doi:10.1136/ijgc-2021-002543.
[4]	FERNANDES F P, CAMBUI R, SOARES J, et al. Cervical carcinoma induces NLRP3 inflammasome activation and IL-1ß release in human peripheral blood monocytes affecting patients' overall survival[J]. Clin Transl Oncol, 2023, 25(11):3277-3286. doi:10.1007/s12094-023-03241-2.
[5]	TANAKA K, MURAKAMI I, MIKAMI M, et al. Reduced expression of gangliosides with GM2-determinant in cervical carcinoma-derived cells after subcutaneous transplantation into nude mice[J]. Hum Cell, 2023, 36(3):1199-1203. doi:10.1007/s13577-023-00864-z.
[6]	YANG H, ZHANG Y, HENG F, et al. Risk prediction model for radiation-induced dermatitis in patients with cervical carcinoma undergoing chemoradiotherapy[J]. Asian Nurs Res(Korean Soc Nurs Sci), 2024, 18(2):178-187. doi:10.1016/j.anr.2024.04.012.
[7]	FURUSAWA A, TAKEKUMA M, MORI K, et al. A randomized phase Ⅲ trial of adjuvant chemotherapy versus concurrent chemoradiotherapy for postoperative cervical cancer:Japanese Gynecologic Oncology Group study(JGOG1082)[J]. Int J Gynecol Cancer, 2021, 31(4):623-626. doi:10.1136/ijgc-2020-002344.
[8]	芦海星, 焦光丽, 刘峰, 等. 影像组学模型对局部中晚期宫颈鳞癌同步放化疗后无进展生存期的预测[J]. 国际医学放射学杂志, 2023, 46(4):402-408.
	LU H X, JIAO G L, LIU F, et al. An radiomics model for progression-free survival prediction after concurrent chemoradiotherapy in patients with locally advanced squamous cervical cancer[J]. Int J Med Radiol, 2023, 46(4):402-408. doi:10.19300/j.2023.L20077.
[9]	段梦瑾, 张路, 田齐, 等. 脑膜瘤的MRI影像组学应用进展[J]. 国际医学放射学杂志, 2023, 46(5):551-555.
	DUAN M J, ZHANG L, TIAN Q, et al. Progress of MRI-based radiomics in meningioma research[J]. Int J Med Radiol, 2023, 46(5):551-555. doi:10.19300/j.2023.Z20580.
[10]	郑明雪, 董江宁, 李翠平, 等. 表观扩散系数联合纹理特征评估宫颈鳞癌分化程度的价值[J]. 实用放射学杂志, 2020, 36(4):592-595,614.
	ZHENG M X, DONG J N, LI C P, et al. The value of ADC values and texture analysis in evaluating the differentiation degree of cervical squamous cell carcinoma[J]. J Pract Radiol, 2020, 36(4):592-595,614. doi:10.3969/j.issn.1002-1671.2020.04.021.
[11]	中国抗癌协会妇科肿瘤专业委员会. 子宫颈癌诊断与治疗指南(2021年版)[J]. 中国癌症杂志, 2021, 31(6):474-489.
	Gynecological Tumor Professional Committee of Chinese Anti-Cancer Association. Guidelines for diagnosis and treatment of cervical cancer(2021 edition)[J]. China Oncology, 2021, 31(6):474-489. doi:10.19401/j.cnki.1007-3639.2021.06.06.
[12]	BHATLA N, DENNY L. FIGO cancer report 2018[J]. Int J Gynaecol Obstet, 2018, 143 Suppl 2:2-3. doi:10.1002/ijgo.12608.
[13]	THERASSE P, ARBUCK S G, EISENHAUER E A, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer,National Cancer Institute of the United States,National Cancer Institute of Canada[J]. J Natl Cancer Inst, 2000, 92(3):205-216. doi:10.1093/jnci/92.3.205.
[14]	ENGINEER R, CHOPRA S, SHUKLA R, et al. Computed tomography-based interstitial brachytherapy for recurrent cervical carcinoma in the vaginal apex[J]. Clin Oncol(R Coll Radiol), 2022, 34(1):e1-e6. doi:10.1016/j.clon.2021.09.012.
[15]	KAUR A, SHARMA Y, KUMAR A, et al. In-vitro antiproliferative efficacy of Abrus precatorius seed extracts on cervical carcinoma[J]. Sci Rep, 2022, 12(1):10226. doi:10.1038/s41598-022-13976-7.
[16]	尹韵清, 张伟, 张彦舫, 等. 影像组学预测肝细胞癌免疫组化标志物的研究进展[J]. 国际医学放射学杂志, 2024, 47(6):719-724.
	YIN Y Q, ZHANG W, ZHANG Y F, et al. Research progress on radiomics in predicting immunohistochemical biomarkers of hepatocellular carcinoma[J]. Int J Med Radiol, 2024, 47(6):719-724. doi:10.19300/j.2024.Z21385.
[17]	田士峰, 刘爱连, 郭妍, 等. 基于扩散峰度成像序列平均扩散峰度图的影像组学鉴别宫颈鳞癌与腺癌的应用[J]. 中国医学影像学杂志, 2021, 29(7):716-720.
	TIAN S F, LIU A L, GUO Y, et al. Application of radiomics methods based on DKI sequence MK map for differentiating squamous cell carcinoma from cervix adenocarcinoma[J]. Chinese Journal of Medical Imaging, 2021, 29(7):716-720. doi:10.3969/j.issn.1005-5185.2021.07.015.
[18]	王艳芳, 肖峰, 张寒菲, 等. DCE-MRI图像纹理分析对乳腺良恶性病变的诊断价值[J]. 武汉大学学报(医学版), 2021, 42(6):935-940.
	WANG Y F, XIAO F, ZHANG H F, et al. Diagnostic value of texture analysis based on DCE-MRI images on benign and malignant breast lesions[J]. Medical Journal of Wuhan University, 2021, 42(6):935-940. doi:10.14188/j.1671-8852.2019.0596.
[19]	佟晶, 卑贵光, 阚杨杨, 等. 多序列MRI影像组学预测局部晚期宫颈鳞癌同步放化疗早期治疗反应的价值[J]. 中华放射学杂志, 2022, 56(11):1223-1229.
	TONG J, BEI G G, KAN Y Y, et al. The value of multi-sequence MRI radiomics in predicting the early response to concurrent chemoradiotherapy for locally advanced cervical squamous cell carcinoma[J]. Chin J Radiol, 2022, 56(11):1223-1229. doi:10.3760/cma.j.cn112149-20211117-01024.

多序列MRI影像组学对局部晚期宫颈鳞癌同步放化疗疗效的预测价值

The predictive value of multi-sequence MRI radiomics in the therapeutic effect of concurrent chemoradiotherapy on locally advanced cervical squamous cell carcinoma

引用本文

RichHTML

PDF (PC)

可视化

摘要/Abstract

使用本文

图/表 8

参考文献 19

相关文章 15

编辑推荐

Metrics

本文评价

[1]	蒋晓涵, 曹杰, 刘丹丹, 薛丹, 郭志国. aMAP评分联合RAR及PIV构建慢性肝病患者肝细胞癌发生风险的预测模型[J]. 天津医药, 2025, 53(1): 42-46.
[2]	梁大敏, 杨正久, 张子萍, 钱静, 毛朝坤. 山萘酚逆转肝癌耐药细胞Bel-7402/5-Fu的作用机制研究[J]. 天津医药, 2024, 52(9): 900-906.
[3]	马佳佳, 张亚苹, 杨斌, 赵美琪, 蒋璐, 黄小玉, 范路畅, 王凤梅. ATOX1通过JAK2/STAT3通路促进肝癌细胞生物学行为的机制探讨[J]. 天津医药, 2024, 52(9): 907-912.
[4]	满祎, 许娅, 何先成, 宋少锋, 刘爱国. 三阴性乳腺癌EGFR、Ki-67、P53及CTC表达与预后的关系研究[J]. 天津医药, 2024, 52(8): 862-867.
[5]	武才鑫, 闫彦, 邓瑗琳, 杜雅敏, 杨振文, 潘晴, 杨帆. 心脏磁共振评估结缔组织病伴重度肺动脉高压的价值[J]. 天津医药, 2024, 52(7): 691-694.
[6]	朱文智, 邱倩, 谭宏宇. 术前焦虑对肝癌患者麻醉苏醒与术后恢复的影响[J]. 天津医药, 2024, 52(7): 762-765.
[7]	朱刚明, 董永德, 朱瑞婷, 谭源满, 陶娟, 刘晓, 陈德成, 杨概. 磁共振弛豫时间定量成像预测乳腺浸润性导管癌分子亚型的价值[J]. 天津医药, 2024, 52(7): 770-774.
[8]	霍晶辰, 王悦, 李华, 邱嵘, 苏景伟, 王卓凡, 杨洁. 系统免疫炎症指数对根治性放疗Ⅲ期肺鳞癌患者长期生存的预测价值[J]. 天津医药, 2024, 52(6): 634-638.
[9]	顾程, 申新宇, 孙静华, 闫赛克, 王海平. 动态MRI成像技术对低位直肠癌保肛术后LARS的评估价值[J]. 天津医药, 2024, 52(6): 653-657.
[10]	张彩蝶, 靳艳, 张德德. 润肺益肾饮对肺癌荷瘤大鼠的抑瘤作用和肿瘤免疫微环境的影响[J]. 天津医药, 2024, 52(4): 362-366.
[11]	吴倩, 王意, 陈念, 周凯, 田昕, 徐晖, 苟小霞. 诱导化疗对鼻咽癌患者免疫功能及炎症指标的影响[J]. 天津医药, 2024, 52(4): 397-402.
[12]	秦爱英, 陈静, 单风晓, 陆翰杰, 武阳. 非小细胞肺癌组织LncRNA MIR503HG、Wnt1表达与患者术后5年内生存的相关性[J]. 天津医药, 2024, 52(4): 403-408.
[13]	徐丹, 刘夏, 钟殿胜. 佐利替尼一线治疗EGFR突变NSCLC伴中枢神经系统转移2例报告[J]. 天津医药, 2024, 52(3): 315-318.
[14]	高光仁, 冯连荣, 付金国, 郭润, 牛和平, 李凤鹏, 张倩玉, 张军. 急性心肌梗死合并胸腔积液患者的心肌损伤特点及对远期预后的影响[J]. 天津医药, 2024, 52(2): 197-200.
[15]	陈慧敏, 贾洪峰, 江婷婷, 贾耀辉. 术中血糖波动和术后胰岛素抵抗对胸腔镜肺癌根治术后老年患者认知功能障碍的影响[J]. 天津医药, 2024, 52(2): 201-205.