多序列磁共振全域直方图分析鉴别子宫肉瘤与变性子宫肌瘤的价值  

The value of multi-sequence magnetic resonance imaging histogram analysis in distinguishing uterine sarcoma from degenerative uterine fibroids

作　　者：邓竞和 刘洋[2] 潘缘蕊 王璐[1] 吕发金[2] DENG Jing-he;LIU Yang;PAN Yuan-rui(State Key Laboratory of Ultrasound in Medicine and Engineering,College of Biomedical Engineering,Chongqing Medical University,Chongqing 400016,China)

机构地区：[1]重庆医科大学超声医学工程国家重点实验室/生物医学工程学院,重庆400016 [2]重庆医科大学附属第一医院放射科,重庆400042

出　　处：《放射学实践》2025年第3期383-389,共7页Radiologic Practice

基　　金：超声医学工程国家重点实验室开放课题基金资助项目(2022KFKT005)。

摘　　要：目的:探究多序列磁共振全域直方图分析鉴别子宫肉瘤与变性子宫肌瘤的价值。方法:回顾性分析子宫肉瘤患者74例和变性子宫肌瘤患者263例的术前多序列MRI图像,利用Firevoxel软件进行灰度直方图分析,提取T_(2)WI、动态对比增强(DCE)及表观扩散系数(ADC)图灰度直方图参数并比较两组疾病的直方图参数差异。对三种序列的直方图参数进行联合,利用二元逻辑回归筛选参数以构建预测模型;绘制受试者操作特征(ROC)曲线,评价各独立参数以及联合模型鉴别子宫肉瘤与变性子宫肌瘤的效能。结果:两组间T_(2)WI直方图参数均值、标准差、峰度值、偏度值、熵值、所有百分位数值、DCE直方图参数特征均值、标准差值、百分位数P25、P50、P75、P95、P99值及ADC直方图参数均值、熵值、所有百分位数值的差异具有统计学意义(P<0.05)。T_(2)WI+DCE+ADC模型具有最大的曲线下面积(AUC=0.907),但根据Delong检验结果其与T_(2)WI+ADC模型(AUC=0.899)在鉴别效能上差异并无统计学意义(P=0.209)。结论:多序列MRI全域直方图分析有助于子宫肉瘤与变性子宫肌瘤的鉴别诊断,T_(2)WI+ADC联合模型有较大的诊断效能提升,并在DCE的参数纳入下具有最高的AUC。Objective:To explore the value of multi-sequence magnetic resonance imaging histogram analysis in differentiating uterine sarcoma from degenerative uterine fibroids.Methods:A retrospective analysis was conducted on preoperative multi-sequence MRI images of 74 patients with uterine sarcoma and 263 patients with degenerative uterine fibroids.Firevoxel software was used to perform grayscale histogram analysis,extracting grayscale histogram parameters of T_(2)WI(T_(2)weighted imaging),DCE(dynamic contrast-enhanced),and ADC(apparent diffusion coefficient)images,and comparing the differences in histogram parameters between the two groups.Combined histogram parameters from the three sequences were used,and binary logistic regression was employed to filter parameters to construct a prediction model.Receiver operating characteristic curve(ROC)curves were plotted to evaluate the effectiveness of independent parameters and joint models in distinguishing uterine sarcoma from degenerative uterine fibroids.Results:Significant differences were observed in T_(2)WI histogram parameter mean,standard deviation,kurtosis,skewness,entropy,all percentile values;DCE histogram parameter characteristic mean,standard deviation,percentile P25,P50,P75,P95,P99 values;ADC histogram parameter mean,entropy,and all percentile values between the two groups(P<0.05).The T_(2)WI+DCE+ADC model exhibited the largest area under the curve(AUC=0.907),but according to the Delong test results,there was no statistically significant difference in discriminatory ability between T_(2)WI+DCE+ADC model and the T_(2)WI+ADC model(AUC=0.899,P=0.209).Conclusion:Multi-sequence magnetic resonance histogram analysis is helpful for the differential diagnosis of uterine sarcoma and degenerative uterine fibroids.The T_(2)WI+ADC combined model demonstrated significant diagnostic efficacy and has the highest AUC when DCE parameters were included.

关 键 词：子宫肉瘤 变性子宫肌瘤 磁共振成像 直方图分析 诊断 鉴别 

分 类 号：R445.2[医药卫生—影像医学与核医学] R737.33[医药卫生—诊断学]