摘要
目的探讨体素内不相干运动(IVIM)多模型参数术前预测子宫内膜癌(EC)风险的意义。资料与方法回顾性分析40例行1.5T MRI检查(含IVIM序列),经手术病理证实为EC患者的影像资料。根据病理类型、病理分级、侵袭程度等分为高风险组(18例)和低风险组(22例)。测量病灶的IVIM多模型参数值,包括单指数模型的标准表观扩散系数(ADC)值,双阶单指数模型的纯扩散参数(D-mono)、灌注参数(D*-mono)、灌注分数(f-mono)值,双阶双指数模型的纯扩散参数(D-Bi)、灌注参数(D*-Bi)、灌注分数(f-Bi)值,以及拉伸指数模型的分布扩散系数(DDC)、拉伸指数(α)值,比较各参数值的差异。采用受试者工作特征(ROC)曲线评估有统计学差异的参数对高、低风险组EC鉴别诊断的效能。结果高风险组的标准ADC值、D-mono值、D-Bi值及DDC值均小于低风险组(P<0.05);两组D*-mono值、f-mono值、D*-Bi值、f-Bi值及α值差异无统计学意义(P>0.05)。标准ADC、D-mono、D-Bi及DDC值预估高风险组EC的曲线下面积分别为0.816、0.788、0.830和0.847,临界值分别为0.608×10^-3 mm^2/s、0.436×10^-3 mm^2/s、0.373×10^-3 mm^2/s和0.738×10^-3 mm^2/s。结论IVIM序列单指数模型的标准ADC值、双阶单指数模型的D-mono值、双阶双指数模型的D-Bi值、拉伸指数模型的DDC值可有效术前预测EC的风险。
Purpose To discuss the significance of the multi-model parameters of intravoxel incoherent motion(IVIM)in the prediction of endometrial carcinoma(EC)risks.Materials and Methods Retrospectively analysis was conducted for the imaging data of 40 cases,who accepted 1.5T MRI examination(including IVIM sequence)and had been proved as EC by surgical pathology.Patients were divided into high-risk group(18 cases)and low-risk group(22 cases)according to pathological types,pathological grades and invasion degrees.The IVIM multi-model parameter values used for lesion measurement concluded the standard ADC value for single exponential model,pure diffusion parameter of two-order single exponential model(D-mono),perfusion parameter(D*-mono),perfusion fraction(f-mono)value,pure diffusion parameters of two-order two-exponential model(D-Bi),perfusion parameter(D*-Bi),perfusion fraction(f-Bi)value,distribution diffusion coefficient(DDC)of tensile exponent model and the heterogeneity of intravoxel diffusion(α)value.The differences of all parameters were compared.The receiver operating characteristic(ROC)curve was applied to evaluate effectiveness of statistically significant parameters in the differential diagnosis of high and low risk groups.Results The standard ADC value,D-mono value,D-Bi value and DDC value of the high-risk group were lower than those of the low-risk group(P<0.05).There was no statistical significance between the two groups in the D*-mono value,f-mono value,D*-Bi value,f-Bi value andαvalue(P>0.05).The area under the curve of EC in high-risk group estimated with standard ADC value,D-mono value,D-Bi value and DDC value were 0.816,0.788,0.830 and 0.847,respectively;with the critical values of 0.608×10^-3 mm^2/s,0.436×10^-3 mm^2/s,0.373×10^-3 mm^2/s and 0.738×10^-3 mm^2/s,respectively.Conclusion The standard ADC value of IVIM sequence single exponential model,D-mono value of two-order single exponential model,D-Bi value of two-order two-exponential model and the DDC value of distribution diffusion coefficient could predict preoperative EC risks effectively.
作者
田士峰
刘爱连
杨伟萍
陈安良
刘静红
朱姣姣
宋清伟
TIAN Shifeng;LIU Ailian;YANG Weiping;CHEN Anliang;LIU Jinghong;ZHU Jiaojiao;SONG Qingwei(Department of Radiology,the First Affiliated Hospital of Dalian Medical University,Dalian 116011,China;不详)
出处
《中国医学影像学杂志》
CSCD
北大核心
2019年第12期948-952,957,共6页
Chinese Journal of Medical Imaging
基金
首都科技领军人才培养工程(Z181100006318003)
关键词
子宫内膜肿瘤
磁共振成像
扩散加权成像
图像处理
计算机辅助
病理学
外科
妇科外科手术
临床方案
危险性评估
Endometrial neoplasms
Magnetic resonance imaging
Diffusion weighted imaging
Image processing,computer-assisted
Pathology,surgical
Gynecologic surgical procedures
Clinical protocols
Risk assessment