摘要
目的探讨MRI检查诊断乳腺髓样癌及纤维腺瘤的价值。材料与方法回顾性分析病理证实的9例乳腺髓样癌及15例纤维腺瘤术前MRI检查资料,分析比较两组病灶的MRI特征、ADCmean值及一阶统计量差异。结果髓样癌组患者年龄大于纤维腺瘤组(t=2.446,P=0.022)。髓样癌组病灶最大径大于纤维腺瘤组(t=2.254,P=0.033);两组间病灶形态、病灶强化特征、病灶坏死及囊变表现、T2WI序列病灶信号特点、病灶扩散加权成像序列信号强度及时间-信号强度曲线类型有差异。两组病灶ADCmean值差异有统计学意义(P<0.05)。经Matlab 2014a平台中提取的病灶一阶统计量中,相关性、能量、均匀性、平均值、标准方差、熵、方差、峰度、IDM 9个定量参数在鉴别诊断中具有统计学意义(P<0.05)。结论病灶MRI特征利于鉴别诊断乳腺髓样癌与纤维腺瘤,病灶的ADCmean值、部分一阶统计量参数在鉴别诊断中具有较高的价值。
Objective:To explore the value of MRI in the diagnosis of breast medullary and fibroadenoma.Materials and Methods:Data of 9 cases of breast medullary cancer and 15 cases of fibroadenoma confirmed by pathology were retrospectively analyzed and compared.Evaluation the differential diagnostic performance of the ADCmean and the first order features of radiomics based on apparent diffusion coefficient(ADC)maps reconstructed in conventional diffusion-weighted images(DWI)of the cases.Results:The age of patients in medullary carcinoma group was older than that of fibroadenoma group(t=2.446,P=0.022).The maximum diameter of lesions in medullary carcinoma group was larger than that in fibroadenoma group(t=2.254,P=0.033).There were differences in lesion morphology,enhancement characteristics,necrosis and cystic degeneration,signal intensity of plain T2WI lesions,signal intensity of DWI lesions and time-signal intensity curve(TIC)types between the two groups.There are statistically difference in ADCmean between the two groups.Through first order features analysis based on ADC map,Correlation,Energy,Homogeneity,Mean,Standard_Deviation,Entropy,Variance,Kurtosis,IDM demonstrated significant difference between breast medullary carcinoma and breast fibro-adenoma group differences.Conclusions:MRI features of lesions are helpful in differentiating medullary breast cancer from fibroadenoma.ADCmean and some first-order features of lesions have high value in the differential diagnosis.
作者
李艳翠
张伶
梁雯
王海丽
彭如臣
LI Yancui;ZHANG Ling;LIANG Wen;WANG Haili;PENG Ruchen(Department of Radiology,Beijing Luhe Hospital Affiliated to Capital Medical University,Beijing 101149,China;Department of Ultrasonography,Beijing Luhe Hospital Affiliated to Capital Medical University,Beijing 101149,China)
出处
《磁共振成像》
CAS
2020年第2期124-128,共5页
Chinese Journal of Magnetic Resonance Imaging
基金
北京市通州区科技计划项目(编号:KJ2019CX012-08)~~
关键词
磁共振成像
乳腺髓样癌
乳腺纤维腺瘤
一阶统计量
magnetic resonance imaging
medullary breast carcinoma
breast fibroadenoma
first-order statistics