磁共振成像2024,Vol.15Issue(5) :148-153,161.DOI:10.12015/issn.1674-8034.2024.05.023

合成MRI在O-RADS MRI 3~5分卵巢附件肿物良恶性鉴别中的价值

Value of synthetic MRI in differential diagnosis of benign and malignant ovarian adnexal lesions with O-RADS MRI score 3-5

李海蛟 曹崑 李晓婷 孙楠 罗瑶 薛珂 杨于昕 孙应实
磁共振成像2024,Vol.15Issue(5) :148-153,161.DOI:10.12015/issn.1674-8034.2024.05.023

合成MRI在O-RADS MRI 3~5分卵巢附件肿物良恶性鉴别中的价值

Value of synthetic MRI in differential diagnosis of benign and malignant ovarian adnexal lesions with O-RADS MRI score 3-5

李海蛟 1曹崑 1李晓婷 1孙楠 1罗瑶 1薛珂 2杨于昕 2孙应实1
扫码查看

作者信息

  • 1. 北京大学肿瘤医院暨北京市肿瘤防治研究所医学影像科,恶性肿瘤发病机制及转化研究教育部重点实验室 北京 100142
  • 2. 北京联影智能影像技术研究院磁共振科研合作部 北京 100089
  • 折叠

摘要

目的 探索合成MRI(synthetic MRI,syMRI)对卵巢-附件影像报告和数据系统(Ovarian-Adnexal Reporting and Data System,O-RADS)MRI 3~5分的卵巢附件占位良恶性鉴别的效能.材料与方法 回顾性分析在2021年8月至2023年6月期间于我院就诊的100例盆腔占位患者病例及影像资料,共计附件肿块126例,所有占位的O-RADS MRI评分均为3~5分.以手术病理或至少1年的随访结果为诊断标准.所有患者均在3.0 T MRI扫描仪上进行盆腔MRI扫描,包括syMRI及扩散加权成像(diffusion-weighted imaging,DWI)序列.在附件区病灶的实性成分的最大层面勾画感兴趣区,以获得syMRI定量参数[T1、质子密度(proton density,PD)、T2*、R2*]及表观扩散系数(apparent diffusion coefficient,ADC)值.采用独立样本t检验或Mann-Whitney U检验比较各参数的差异,通过logistic回归分析建立syMRI及syMRI+ADC诊断模型,采用受试者工作特征(receiver operating characteristic,ROC)曲线比较各参数及模型的诊断效能,DeLong检验比较各模型ROC曲线下面积(area under the curve,AUC)的差异.结果 入组100例患者共126个病灶,其中良性55例,恶性71例.T1、T2*、R2*及ADC值在两组间差异有统计学意义(P<0.05),其鉴别卵巢附件良恶性病变的AUC分别为 0.739[95%置信区间(confidence interval,CI):0.652~0.826]、0.780(95%CI:0.698~0.862)、0.783(95%CI:0.699~0.866)及0.674(95%CI:0.576~0.772).syMRI及syMRI+ADC模型的AUC分别为 0.860(95%CI:0.791~0.929)及 0.879(95%CI:0.818~0.940),二者之间差异无统计学意义,均高于ADC值(P<0.05).结论 syMRI在鉴别O-RADS MRI 3~5分卵巢附件病变的良恶性中具有很好的效能.

Abstract

Objective:To investigate the diagnostic efficacy of synthetic MRI(syMRI)in differentiation of benign and malignant ovarian adnexal lesions with Ovarian-Adnexal Reporting and Data System(O-RADS)MRI score 3-5.Materials and Methods:Totally 100 patients with 126 ovarian adnexal lesions scored 3-5 according to O-RADS MRI in our hospital from August 2021 to June 2023 were retrospectively enrolled.The diagnosis was confirmed via pathological examination or one-year follow-up.All patients underwent pelvic syMRI and diffusion-weighted imaging(DWI)on a 3.0 T MR scanner.Regions of interest(ROIs)were placed on the largest slice of the solid part of adnexal lesion and avoid the cystic or necrotic areas.Quantitative parameters of syMRI[T1,proton density(PD),T2*,R2*]and apparent diffusion coefficient(ADC)values were calculated.Independent samples t test and Mann-Whitney U test were utilized to compare the differences of quantitative parameters between benign and malignant lesions.Two models,the syMRI model and the syMRI+ADC model,were constructed using logistic regression analysis.Receiver operating characteristic(ROC)analysis was used to evaluate the diagnostic efficacy of the individual quantitative parameters and diagnostic models.DeLong test was employed to compare the difference of area under the curve(AUC).Results:Among 126 ovarian adnexal lesions,55 lesions were benign,and 71 lesions were malignant.T1,T2*,R2*,and ADC showed significant difference between groups(all P<0.05)and the AUCs of these parameters in differentiating benign and malignant adnexal lesions were 0.739[95%confidence interval(CI):0.652-0.826],0.780(95%CI:0.698-0.862),0.783(95%CI:0.699-0.866),and 0.674(95%CI:0.576-0.772),respectively.The AUCs of syMRI and syMRI+ADC models were 0.860(95%CI:0.791-0.929)and 0.879(95%CI:0.818-0.940),respectively.The AUCs of the two models showed no statistical difference,and both of them were higher than that of ADC(all P<0.05).Conclusions:The syMRI proved to be valuable in differential diagnosis of benign and malignant ovarian adnexal lesions with O-RADS MRI score 3-5.

关键词

卵巢肿瘤/磁共振成像/合成MRI/卵巢-附件影像报告和数据系统/鉴别诊断

Key words

ovarian neoplasms/magnetic resonance imaging/synthetic magnetic resonance imaging/Ovarian-Adnexal Reporting and Data System/differential diagnosis

引用本文复制引用

基金项目

国家重点研发计划(2023YFC3402805)

北京大学肿瘤医院科学研究基金(2022-20)

北京大学肿瘤医院科学研究基金(XKFZ2403)

出版年

2024
磁共振成像
中国医院协会 首都医科大学附属北京天坛医院

磁共振成像

CSTPCDCSCD北大核心
影响因子:1.38
ISSN:1674-8034
参考文献量39
段落导航相关论文