MR动态增强及减影成像对乳腺良恶性疾病的诊断价值

目的探讨MR动态增强成像技术所获得的参数及减影成像在乳腺良、恶性疾病诊断及鉴别诊断中的价值。资料与方法经病理证实乳腺病变50例，共计58个病灶。采用快速小角度激发扰相梯度回波序列（FLASH）行动态增强扫描。观察：时间一信号强度曲线类型，早期增强率，增强后病灶形态学特征及间接征象，于3D减影最大强度投影（MIP）图计数肿瘤血管数目。结果50例58个病灶经病理证实，其中恶性32个，良性26个。良性病变中时间一信号强度曲线57．7％为单向型，38．5％为平台型。恶性病变中71．9％为流出型。良、恶性病变的时间一信号强度曲线类型比较，差异有统计学意义（X^2=32．365，P＜0．05）。良性肿瘤中95％边缘光滑，55％均匀强化；恶性肿瘤中93．8％的形态不规则，78．1％有毛刺和分叶，81．3％不均匀强化。良、恶性肿瘤的形态特点差异有统计学意义（X^2=44．25，P＜0．05）。良性病变中73．1％无肿瘤血管：恶性病变中68．8％有多支肿瘤血管，良、恶性病变肿瘤血管计数差异有统计学意义（X^2=32．288，P＜0．05）。结论乳腺增强MRI及减影成像能够提供清晰、丰富的形态学信息，时间一信号强度曲线反映了病变的血流动力学信息，两者的结合对乳腺良、恶性疾病的正确诊断和鉴别诊断具有重要价值。     

IVIM-DWI辅助DCE-MRI对良、恶性乳腺非肿块样强化病变的鉴别诊断价值

【摘要】目的：探讨体素内不相关运动扩散加权成像（IVIM-DWI）对磁共振动态增强扫描（DCE-MRI）鉴别良、恶性乳腺非肿块样强化病变的辅助诊断效能。方法：回顾性分析56例呈非肿块样强化的乳腺病变的影像资料,分析病变在DCE-MRI上的特征和定量参数［(病变分布特征、内部强化特点、TIC曲线、最大增强斜率(MSI)］及IVIM-DWI定量参数（f值、D值、D*值）并进行良、恶性组间比较;建立单独应用DCE-MRI及DCE-MRI与IVIM-DWI联合应用的Logistic回归模型,评价其对良、恶性病变的鉴别诊断效能。结果:病变的分布特征、强化特点及TIC曲线类型在良、恶性组间的差异具有统计学意义（P<0.05）;而MSI（恶性组：2.57、1.89;良性组：1.92、1.43）的组间差异无统计学意义（P>0.05）;恶性组的f值（8.13±2.64）和D*值（6.96±2.89）高于良性组（分别为6.92±2.56和 6.83±2.58）,D值低于良性组（恶性组：1.05±0.31;良性组：1.27±0.26）,其中D*和D值在两组间的差异有统计学意义（P<0.05）。两种回归模型诊断乳腺良、恶性病变的敏感度、特异度、符合率、阳性预测值和阴性预测值分别为88.24%、68.42%、81.13%、79.49%、80.75%及91.18%、73.68%、84.91%、83.35%、85.24%。结论：IVIM-DWI作为一种辅助检查方法能提高DCE-MRI检查对乳腺非肿块样强化病变的鉴别诊断效能,尤其是诊断特异性。     

动态增强MRI鉴别乳腺良恶性病变的价值

目的：探讨动态增强MRI对乳腺良恶性病变的鉴别诊断价值。方法：对147例经手术病理证实的乳腺疾病患者行MRI平扫和动态增强检查。分析比较乳腺良恶性病变的MRI征象。包括形态学、内部信号及其动态增强特征等，计算病灶峰值增强率Emax，峰值时间Tmax，最大强化速率Slope max，并根据病变最高强化区ROI测量值，绘制病灶的时间-信号强度曲线。结果：147例患者共计病灶152个。其中恶性87个，良性65个。分析比较病变的形态学及动态增强表现，发现乳腺良性病变多呈类圆形或分叶状．边缘光整，无毛刺、边界多清晰，强化均匀或伴有内部分隔等，动态强化参数Emax多大于90％，Tmax多大于240s。Slopemax。多大于3％／s；恶性病变则多为不规则形。边缘可伴毛刺，与周围结构分界不清，强化不均匀，病灶常呈环形强化，导管样强化，Emax多大于90％，Tmax多大于240s。而Slopemax多大于3％／s。时间-信号曲线类型恶性病变多表现为Ⅱ或Ⅲ型，良性病变则多为Ⅰ或Ⅳ型。结论：乳腺MRI不仅提供了病灶丰富的形态学信息，而且还通过动态增强进一步揭示病变的血流动力学特征，从而有助于对乳腺病变作出准确的定性诊断，更有效地鉴别诊断乳腺良恶性病变。     

动态增强MRI对乳腺环形强化病灶的诊断价值

目的 研究乳腺内环形强化病变的MRI表现及病理特征,探讨动态增强MRI对乳腺环形强化病灶的诊断价值.方法 回顾性分析23例乳腺MRI上环形强化病例,所有病例均经手术、病理或临床治疗证实.结果 23例乳腺环形强化病变中,恶性病灶17例,良性病灶6例,病灶增强形态特征、早期增强率等及时间-信号强度时间曲线在良恶性乳腺病变中差异均有显著性意义(P<0.05).结论 乳腺环形强化高度提示恶性肿瘤,动态增强MRI通过分析病灶形态学特征和强化方式,可以有效地鉴别诊断乳腺环形强化病灶的良恶性.     

乳腺良恶性病变的动态增强MR信号强度分析

目的:探讨动态增强MB信号强度在鉴别良恶性病变中的意义.方法:72例经手术及病理证实患者,观察和测量其MR信号强度,比较良恶件病变的MR平扫信号强度、动态增强时间-信号强度曲线和早期信号增强率、峰值信号增强率.结果:单以信号强度来说,平扫对于良恶性病变无统计学意义;动态增强时间-信号强度曲线中,良性病变以Ⅰ型曲线为主,占78.9%,而以乳腺痛为代表的恶性病变则以Ⅲ型曲线为主,占79.4%.乳腺癌诊断的敏感性为94.11%,特异性为79.41%,假阳性率为0.08%;早期信号强度增强率对于良恶性病变的鉴别具有统计学意义.结论:动态增强MR信号强度对于乳腺良恶性病变的诊断和鉴别诊断具有很高的临床价值,但本研究结果没有突破,对于定性困难的病灶,还需结合病灶形态学和临床表现及其他检查手段综合分析,以提高对乳腺疾病诊断的正确率.     

动态增强磁共振联合扩散加权成像技术对乳腺环形强化病变的诊断应用价值研究

目的 探讨动态增强MRI(DCE-MRI)联合扩散加权成像(DWI)技术对乳腺环形强化病变的诊断应用价值.方法 搜集经病理证实的乳腺环形强化病变28例,共30个病灶(良性17个,恶性13个),术前行磁共振扫描.统计病变的动态增强MRI征象(形态学特点、时间-信号曲线)及扩散加权成像参数(表观扩散系数值),并运用x2检验及t检验分析良恶性组差异;采用二分类Logistic回归分析获得与乳腺环形强化病变良恶性相关的危险因素.将DCE-MRI及DWI纳入不同的诊断模型,并评价其诊断准确性.结果 动态增强MRI征象(包括形态、边缘、周围供血血管及时间-信号曲线)及扩散加权成像参数在乳腺良、恶性环形强化病变中有显著性差异(P＜0.05);且经二分类Logistic回归分析提示这些征象及参数均为鉴别良恶性乳腺环形强化病变有意义的参数.当表观扩散系数(ADC)值取1.05×10-3mm2/s作为乳腺恶性环形强化病变的诊断阈值时,敏感性及特异性分别为92.3％、70.6％(P=0.001).DCE-MRI联合DWI可进一步提高乳腺环形强化病变的诊断准确性,其敏感性、特异性及符合率分别为94.1％、76.9％及86.7％.结论 动态增强MRI联合DWI技术,可明显提高乳腺环形强化病变的定性诊断价值.     

磁共振动态增强、扩散加权成像联合应用对乳腺病变的诊断价值

目的：探讨单一磁共振动态增强扫描诊断试验、扩散加权成像诊断试验及其联合应用对乳腺病变定性诊断的敏感性、特异性和阳性似然比、阴性似然比,比较其诊断效能。方法：对临床拟诊肿块的患者37例,同时进行动态增强扫描和扩散加权成像检查,均获得手术和病理证实,其中良性病灶18个,恶性病灶19个。对病变的边缘、形态特征、动态增强表现及时间一信号强度曲线采用评分法对病变性质分恶性、可疑恶性及良性三组进行判断。参照动态增强病变位置确定扩散图像病变所在,描记扩散图像上病变的感兴趣区,由软件计算获得表观扩散系数（ADC）值。对获取数据进行统计分析,采用t检验统计学方法进行良性和恶性ADC值比较。联合动态增强扫描和ADC值,采用评分法根据积分情况进行综合定性诊断。比较动态增强扫描、DWI ADC值及联合应用对乳腺病变定性诊断效能。结果：动态增强扫描（病灶边缘、形态学表现结合时间-信号强度曲线）诊断乳腺病变的敏感性、特异性和阳性似然比、阴性似然比分别为89．5％、72．2％和3．221、0．146。良性病变组ADC值1．474±0．441（×100^-3mm^2／s）,恶性病变组ADC值1．082±0．160（×10^-3mm^2／s）,两者间有显著统计学差异（P=0．002,〈0．05）。ADC值诊断敏感性、特异性和阳性似然比、阴性似然比分别为94．7％、66．7％和2．842、0．079。动态增强扫描和DWI-ADC值联合诊断的敏感性、特异性和阳性似然比、阴性似然比分别为94．7％、83．3％和5．684、0．063。结论：磁共振动态增强、扩散加权成像联合应用对乳腺病变的定性诊断敏感性、特异性、阳性似然比、阴性似然比均较单一动态增强扫描或扩散成像诊断效能强。     

MR动态增强减影技术对乳腺良恶性病灶鉴别的价值

目的:评价MR动态增强减影技术对乳腺良恶性病灶鉴别的价值。方法:收集进行乳腺MR动态增强减影扫描并有明确病理结果的56例67个病灶作为研究对象进行回顾性分析。结果:乳腺良恶性病灶在MR动态增强形态、强化时间-信号强度曲线、强化峰值时间、血管分布形态、腋窝淋巴结五个指标上表现不同。结论:MR动态增强减影技术对乳腺良恶性病灶具有较高的鉴别诊断价值。     

MR 动态增强联合扩散加权成像对壶腹区良恶性病变的鉴别

目的：研究M R动态增强联合扩散加权成像（DWI）在鉴别壶腹区良恶性病变的价值。方法回顾性分析43例胆总管下段狭窄患者的M R动态增强及DWI的数据。其中包括32例恶性病变和11例慢性炎症。1位影像医生对壶腹周围良恶性病变的M R动态增强信号强度及DWI信号进行分析,另外2位影像医生对壶腹周围病变的M R动态增强影像以及M R动态增强联合DWI影像进行评估。应用 Logistic回归分析比较灵敏度及特异性。结果壶腹周围良恶性病变MR动态增强表现差异无统计学意义;DWI影像中,壶腹周围癌比炎症更多地表现为高信号,表观扩散系数（ADC）图表现为低信号（P＜0．001）。2位读片者在结合DWI影像后对恶性壶腹周围病变的诊断灵敏度均有提高,分别从84．4％提高到96．9％和从87．7％提高到96．6％。结论 M R动态增强联合DWI可提高鉴别壶腹周围区良恶性狭窄的诊断准确率。     

磁共振动态增强联合扩散加权成像对乳腺良恶性病变的诊断价值

【摘要】目的：探讨磁共振动态增强（DCE-MRI）及扩散加权成像（DWI）对乳腺良恶性病变的定性诊断价值。方法：回顾性分析经手术病理证实的122例乳腺病变患者的临床和影像学资料，所有病例术前行双乳MRI检查。分析病灶的形状、边界、强化方式、早期强化率（EER）、时间-信号强度曲线（TIC）及表观扩散系数（ADC）值，参照Fischer评分标准对影像表现进行评分，根据乳腺影像报告与数据系统第5版（BI-RADS）进行分类诊断。与病理结果对照，计算DCE-MRI、DWI、DCE联合DWI对乳腺病变的诊断敏感度、特异度和符合率，并采用ROC曲线分析其诊断效能。结果：122例中恶性80例，良性42例。DCE-MRI诊断敏感度为87.5%，特异度87.5%，符合率86.9%。以恶性病变ADC值的95%可信区间上限1.225×10-3mm2/s作为鉴别诊断阈值，敏感度为85.7%，特异度78.9%，符合率83.6%。DCE联合DWI的诊断敏感度达93.8%，特异度90.5%，符合率92.6%。DCE联合DWI鉴别乳腺良恶性病变的AUC（0.915）高于单独诊断（0.866，0.855）。结论：DCE-MRI鉴别乳腺良恶性病变的诊断效能较高，DWI可提供辅助诊断信息，DCE与DWI联合诊断能明显提高对乳腺病变的术前定性诊断准确性。     

磁共振纹理动态特征鉴别乳腺良恶性肿块

目的:提取乳腺病灶的时空变化特征作为新的DCE-MRI标记(称为纹理动态特征)并证明其鉴剐良恶性肿块的能力.方法:回顾性分析52个乳腺肿块,其中恶性肿瘤30个,良性肿块22个,提取并对动态特征信号强度特征、纹理特征、形态特征、边缘特征进行分组.为了更好评估这些特征,采用不同的特征类建立分组模型,计算正确率,敏感度,特异性及曲线下面积(AUC).结果:结合纹理动态特征所建立的良恶性肿瘤分类器具有最大的AUC=0.94,准确率90％,敏感度92％,特异性85％,优于其他各组分类器,与信号强度特征所建立的模型差异有统计学意义(AUC=0.80,P＜0.05).结论:磁共振纹理动态特征有助于鉴别良恶性肿块,甚至优于临床上最流行的DCE MRI标记信号强度动态特征.     

常规MRI及DCE-MRI在卵巢肿瘤及肿瘤样病变诊断中的应用

目的:提高常规MRI及DCE-MRI在卵巢肿瘤及肿瘤样病变的诊断准确率。方法:回顾性分析44例我院收治并经手术病理证实为卵巢肿瘤及肿瘤样病变的常规MRI及DCE-MRI影像表现。平扫参考指标:病灶大小(最大径)、形状(规则/不规则)、有无包膜、囊壁/分隔厚度、实性成分比例、T2WI信号、ADC值、盆腔积液;DCE-MRI参考Ktrans、Kep以及Ve值。结果:MRI平扫显示良、恶性肿瘤的病灶大小、形状、实性成分比例、囊壁/分隔厚度、T2WI信号混杂程度、盆腔积液量及ADC值差异有统计学意义(P<0.05);上述7项中ADC值参考价值最大,ADC取1.09×10^-3)mm^2/s为良恶性肿瘤阈值,小于阈值,考虑恶性。DEC-MRI良性肿瘤实性部分Ktrans、Kep以及Ve值显著低于恶性肿瘤,差异有统计学意义(P<0.05),Ktrans值参考价值最大,Ktrans值取0.091/min为良恶性肿瘤阈值,大于阈值,考虑恶性。分别以ADC值及Ktrans值敏感度、特异性、准确度、阳性预测值及阴性预测值代表MRI平扫及DCE-MRI各项指标,得出常规MRI敏感度85.7%、特异性91.3%、准确度88.6%、阳性预测值90.0%、阴性预测值87.5%;DCE-MRI上述各项指标为72.7%、81.8%、77.3%、80.0%、75.0%;常规MRI联合DCE-MRI检查敏感度95.2%,特异性100.0%,准确度97.7%,阳性预测值100.0%,阴性预测值95.8%。结论:常规MRI联合DCE-MRI用于卵巢肿瘤及肿瘤样疾病诊断及鉴别诊断能够提高诊断准确率。     

磁共振扩散加权成像及动态增强MRI在乳腺病变中的应用价值

目的 探讨月经周期对正常乳腺实质磁共振扩散加权成像(DWI)表观扩散系数(ADC)值的影响,并评价动态增强MRI(DCE-MRI)及DWI在乳腺病变鉴别诊断中的价值。方法 选取健康女性志愿者34例、乳腺病变患者96例,所有病例均经外科手术或针吸病理证实,其中良性病变者36例,共44个病灶;恶性病变者60例,共70个病灶...     

基于MR早期动态增强的影像组学标签鉴别乳腺良恶性病变的价值

目的:探讨基于MR早期动态增强的影像组学标签鉴别乳腺良恶性病变的价值。方法:回顾性搜集通过乳腺动态对比增强MRI(DCE-MRI)检查,发现乳腺结节或肿块的144例患者(146个病变),146个病变按照样本量7:3随机抽样选取良性病变与恶性病变(102个作为训练组,44个作为验证组)。所有病例基于病变的三维图像对影像组学特征进行提取,然后采用Lasso logistic回归模型进行特征降维及筛选,以建立影像组学标签。采用ROC曲线对乳腺良恶性病变的诊断效能进行评价。结果:3个主要的影像组学特征与乳腺良恶性病变的鉴别诊断相关(P=0.0005)。建立的影像组学标签对鉴别乳腺良恶性病变具有较高的诊断效能,在训练组中的ROC曲线下面积(AUC)为0.909(95%CI:0.843~0.975),在验证组中的AUC为0.877(95%CI:0.743~1.000)。结论:基于MR早期动态增强构建的影像组学标签对乳腺良恶性病变具有较高的鉴别诊断效能,可辅助临床进行更精准的良恶性分层,为临床治疗方案的制订提供参考。     

乳腺磁共振扩散加权成像的应用

目的：探讨乳腺磁共振扩散加权成像（DWI）检查的可行性,并重点探讨影响DWI图像质量的技术参数。方法：使用GE1．5T磁共振扫描仪及阵列线圈对32例乳腺疾病患者行常规SE序列扫描,其中19例为良性肿瘤,5例炎性病变,6例恶性肿瘤．均经手术及病理证实;另硅胶置入2例。使用体线圈行DWI序列扫描,采用全方位扩散梯度及5个b值扫描。DWI总的扫描时间40s。结果：在DWI序列扫描中,良性和恶性肿瘤均为高信号,计算ADC值可鉴别良性和恶性肿瘤．通过各种扫描参数的合理匹配,可使图像质量的信噪比达到最佳,并减少图像的几何变形。结论：DWI对于检查乳腺病变是一种快速可行并行之有效的技术。     

Magnetic resonance elastography of the breast: correlation of signal intensity data with viscoelastic properties

OBJECTIVES: We sought to investigate the potential value of magnetic resonance (MR) elastography to improve the differentiation between benign and malignant tumors. MATERIAL AND METHODS: Measurements of 5 patients with 6 malignant lesions, 11 patients with benign lesions, and 4 patients with no lesions at all were performed at 1.5 Tesla. After breast MR imaging, MR elastography was performed as a targeted measurement. Low-frequency mechanical waves (65 Hz) were transmitted into the breast tissue using an oscillator and were displayed by means of a MR sequence within the phase of the MR image. After reconstruction, the viscoelastic information was correlated with the signal intensity and morphology data. RESULTS: All examinations were technically successful realized in approximately 25 minutes. Malignant tumors documented higher values of elasticity than benign corresponding with signal intensity and morphologic data. CONCLUSION: A good separation exists between benign and malignant lesions in elasticity, corresponding with specific signal intensity and morphologic data. Further clinical studies with a larger number of patients are needed for extended validation.     

Dynamic contrast enhanced MRI in the differential diagnosis of soft tissue tumors

PURPOSE: The value of the dynamic contrast enhanced-magnetic resonance imaging (DCE-MRI) in differentiating benign and malignant soft tissue tumors was investigated. MATERIALS AND METHODS: Turbo FLASH DCE-MRI was performed on 22 subjects (2-74 years) with soft tissue tumors. Enhancement in the first min (E(max/1)), second min (E(max/2)) and maximum peak enhancement (E(max)), and steepest slope were calculated. Discriminant analyses were performed to reveal parametric differences of benign and malignant lesions. RESULTS: Diagnosis of benign (N = 10) tumors were hemangioma (n = 3), neurogenic tumor (n = 3) lipoma (n = 2), giant cell tumor (n = 1) and desmoid (n = 1), whereas malignant lesions (N = 12) were classified as liposarcoma (n = 5), malignant fibrous histiocytoma (n = 5) and synovial sarcoma (n = 2). For malignant lesions E(max/1) was 65-198%, E(max/2) was 65-145%, E(max) was 78-198%, and steepest slope was 1.45-4.06. For benign lesions these values were 4-98%, 5-105%, 7-125% and 0.67-2.57, respectively. To determine the relation between the variables analysed, Pearson correlation coefficients were calculated. E(max) was found to be highly correlated with other variables (rxy > 0.86, P < 0.0001). Consequently, this variable was excluded from the discriminant analysis. In order to determine discrimination of malignant and benign tumors using E(max/1), E(max/2,) and steepest slope of the enhancement curve logistic regression was applied to the above mentioned data. When combined these parameters had a 95.5% of overall accuracy in classifying benign and malignant lesions (P = 0.004). CONCLUSION: DCE-MRI parameters that thought to be the surrogate markers of tumoral microcirculation and tissue perfusion provides a specific preoperative diagnosis. Dynamic imaging parameters are therefore advocated for monitoring the effect of chemotherapy in soft tissue tumors.     

MRI动态增强定量参数对肝细胞癌和肝良性肿瘤鉴别诊断的价值

目的：研究磁共振动态增强(DCE-MRI)定量参数在肝细胞癌和肝良性肿瘤鉴别的意义,探索定量参数对肝脏占位性病变良恶性诊断的界值及其诊断效能。方法对存在肝占位的25例患者进行 DCE-MRI 扫描,经病理证实或通过随访确诊肝细胞癌和肝良性肿瘤共28个病灶,利用 tissue4D 后处理软件测定病灶及周围正常肝组织的定量参数值(Ktrans 、Kep 、Ve 、iAUC),利用独立样本 t 检验比较肝细胞癌组、良性组病灶与周围正常肝组织的定量参数,同时比较2组病灶间的定量参数值,并通过绘制受试者工作特征曲线(ROC)来确定肝细胞癌和肝良性肿瘤间定量参数的最佳界值及其诊断效能。结果良性组、肝细胞癌组病灶的各定量参数均值大于周围正常肝组织,其中良性组 Ktrans 、iAUC 差异有统计学意义(P <0.05);肝细胞癌组 Ktrans 、Kep 、iAUC 差异有统计学意义(P <0.05);肝细胞癌组和良性组比较,肝细胞癌组病灶的各定量参数均值大于良性组,Ktrans 、Kep 及 iAUC 的差异有统计学意义(P <0.05)。其中 Ktrans =0.215 min-1时,敏感度为81.3%,特异度为66.7%,诊断效能大于 Kep (界值为0.477)、iAUC(界值为24.706)。结论DCE-MRI 定量参数能为肝细胞癌和良性肿瘤的鉴别诊断提供参考。     

A computerized global MR image feature analysis scheme to assist diagnosis of breast cancer: a preliminary assessment

Objectives

To develop a new computer-aided detection scheme to compute a global kinetic image feature from the dynamic contrast enhanced breast magnetic resonance imaging (DCE-MRI) and test the feasibility of using the computerized results for assisting classification between the DCE-MRI examinations associated with malignant and benign tumors.

Materials and Methods

The scheme registers sequential images acquired from each DCE-MRI examination, segments breast areas on all images, searches for a fraction of voxels that have higher contrast enhancement values and computes an average contrast enhancement value of selected voxels. Combination of the maximum contrast enhancement values computed from two post-contrast series in one of two breasts is applied to predict the likelihood of the examination being positive for breast cancer. The scheme performance was evaluated when applying to a retrospectively collected database including 80 malignant and 50 benign cases.

Results

In each of 91% of malignant cases and 66% of benign cases, the average contrast enhancement value computed from the top 0.43% of voxels is higher in the breast depicted suspicious lesions as compared to another negative (lesion-free) breast. In classifying between malignant and benign cases, using the computed image feature achieved an area under a receiver operating characteristic curve of 0.839 with 95% confidence interval of [0.762, 0.898].

Conclusions

We demonstrated that the global contrast enhancement feature of DCE-MRI can be relatively easily and robustly computed without accurate breast tumor detection and segmentation. This global feature provides supplementary information and a higher discriminatory power in assisting diagnosis of breast cancer.     

Multi-section magnetic susceptibility perfusion echo-planar imaging of the breast

PURPOSE: To examine the diagnostic value of multi-section magnetic susceptibility perfusion echo-planar imaging (perfusion EPI) in patients with breast tumors prior to T1-weighted dynamic MRI. MATERIALS AND METHODS: MR imaging was performed in 75 patients with pathologically proved breast tumors using a 1.5 Tesla MR unit (MAGNETOM Symphony, Siemens Medical Solutions, Erlangen, Germany). Perfusion EPI was carried out before, during, and after the bolus injection of 0.1 mmol Gd-DTPA/kg. Two patients had two carcinomas in the same breast, one patient had both a benign and a malignant lesion in the same breast, and two patients had lesions in both breasts. Histopathological diagnosis was non-invasive ductal carcinoma in 9, invasive carcinoma in 49, and benign lesion in 22. The first-pass signal intensity loss of the lesions was calculated by perfusion EPI. RESULTS: Fifty-one of 58 carcinomas but only 4 of 22 benign lesions had a signal intensity loss of 20% or more during the first pass, for a sensitivity of 88% and specificity of 82% CONCLUSION: Perfusion EPI can be used as a useful diagnostic tool for differentiation between benign and malignant lesions. It is also thought to be a promising method for diagnosing multifocal breast lesions.