吉林大学学报(医学版) ›› 2016, Vol. 42 ›› Issue (01): 159-163.doi: 10.13481/j.1671-587x.20160133

• 影像学 • 上一篇    下一篇

ADC和rADC值对乳腺良恶性病变的诊断效能

鲁文力, 欧阳祖彬, 李信友, 石军, 张志伟   

  1. 重庆医科大学附属第一医院放射科, 重庆 400016
  • 收稿日期:2015-04-28 发布日期:2016-01-26
  • 通讯作者: 欧阳祖彬,副主任医师(Tel:023-89012227,E-mail:903459875@qq.com) E-mail:903459875@qq.com
  • 作者简介:鲁文力(1988-),女,重庆市人,医学硕士,主要从事神经系统和骨骼影像学方面的研究。
  • 基金资助:

    国家临床重点专科建设项目资助课题(国卫办医函[2013]544号)

Diagnostic efficiency of ADC and rADC on benign and malignant breast lesions in diffusion weighted imaging

LU Wenli, OUYANG Zubin, LI Xinyou, SHI Jun, ZHANG Zhiwei   

  1. Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China
  • Received:2015-04-28 Published:2016-01-26

摘要:

目的: 观察乳腺扩散加权成像(DWI)定量参数表观扩散系数(ADC)及相对表观扩散系数(rADC)值对鉴别乳腺良恶性病变的诊断效能。 方法: 选择经穿刺或手术病理证实并行乳腺MRI检查的女性患者123例,行轴位DWI、矢状位T2WI及动态增强MRI扫描(DCE-MRI)。以DCE-MRI图像为参考,在DWI图上分别于病灶内、病灶邻近正常腺体区域放置感兴趣区(ROI),记录相应区域ADC值,并计算标化后rADC值。比较乳腺良性和恶性病变的平均ADC值及rADC值。根据受试者工作特征(ROC)曲线确定ADC和rADC鉴别乳腺良性和恶性病变的最佳诊断阈值;比较ROC曲线下面积(AUC),评价两者对乳腺良恶性病变的诊断效能。 结果: 134个乳腺病灶中,84个恶性病灶和50个良性病灶平均ADC值分别为(1.04±0.26)×10-3和(1.60±0.40)×10-3mm2·s-1,正常腺体134个ROI平均ADC值为(1.89±0.31)×10-3 mm2·s-1,恶性病变ADC值明显低于良性病变和正常腺体(P<0.01)。恶性病变和良性病变平均rADC值分别为0.59±0.18和0.83±0.30,恶性病变rADC值低于良性病变(P<0.01)。选择ADC阈值为1.22×10-3 mm2·s-1,其敏感度和特异度分别为86.0%和84.5%,AUC为0.89。选择rADC阈值为0.67,其敏感度和特异度分别为74.0%和75.0%,AUC为0.79。标化后rADC的AUC低于标化前ADC(P<0.05)。 结论: ADC和rADC 值的定量分析是鉴别诊断乳腺良性和恶性病变的有效手段,标化后rADC 值未能提升ADC值的诊断效能。

关键词: 乳腺病变, 扩散加权成像, 表观扩散系数, 相对表观扩散系数, 诊断效能

Abstract:

Objective: To investigate the diagnostic efficiency of the apparent diffusion coefficient(ADC)and the relative apparent diffusion coefficient (rADC)in the differentiation between benign and malignant breast lesions in diffusion weighed imaging(DWI).Methods: One hundred twenty-three females with suspected breast lesions who underwent breast MRI examination and proved by pathology and surgery were included in this study. The axial DWI,sagittal T2WI and dynamic contrast enhanced (DCE)scan were performed. With reference to the location and size of lesions on the DCE image,the regions of interest (ROI)were placed in the lesions and their adjacent normal gland regions (more than 2 cm to the lesion). The ADC values of corresponding regions were recorded,and the standardized rADC values were calculated. The mean values of ADC and rADC between benign and malignant breast lesions were compared. The optimal threshold values of ADC and rADC in differentiation between benign and malignant lesions were defined based on receiver operating characteristic (ROC)curve. Their diagnostic efficiency in breast lesions were assessed by comparing the area under curve(AUC). Results: Among the 134 lesions,the mean ADC values of 84 malignant lesions and 50 benign lesions were (1.04±0.26)×10-3 and (1.60±0.40)×10-3 mm2·s-1,respectively. The mean ADC value of 134 ROI in the normal,glands was (1.89±0.31)×10-3 mm2·s-1. The ADC value of malignant lesions was significantly lower than those of benign lesion and normal gland tissues(P<0.01). The mean rADC values of malignant lesions and benign lesion were 0.59±0.18 and 0.83±0.30,respectively;the rADC value of malignant lesion was lower than that of benign lesion (P<0.01).When the ADC cutoff value of 1.22×10-3 mm2·s-1 was selected,its sensitivity and specificity were 86.0% and 84.50%,respectively;its AUC was 0.89. While the rADC cutoff value of 0.80 was selected,its sensitivity and specificity were 74.0% and 75.0%,respectively;its AUC was 0.80. The AUC of standardized rADC was lower than that of unstandardized ADC(P<0.05).Conclusion: Quantitative analysis of ADC and rADC values is an efficient method for differentiation between benign and malignant breast lesions.The standardized rADC value does not improve the diagnostic efficiency of ADC value.

Key words: breast lesion, diffusion weighted imaging, apparent diffusion coefficient, relative apparent diffusion coefficient, diagnostic efficiency

中图分类号: 

  • R737.9