Dynamic Contrast-enhanced MRI Of Breast Tumor

Breast cancer is one of the most common cancers in women. Traditional imaging techniques (mammography and ultrasound) have high value in detection and diagnosis of breast cancer, but they have limitation, too. With the development of MRI techniques and the clinical application of fast imaging, diagnosis of breast tumor arrives a higher level. Dynamic enhanced MRI provides not only morphologic features but also kinetic characteristics of breast tumor with high sensitivity and specificity. However, because the foci and parameters were different in different researches, there has never been a technical standard or diagnostic criteria for breast MRI yet.1. PURPOSETo investigate morphologic and kinetic characteristics of breast tumor. To make different diagnostic criteria for benign and malignant tumors based on their different features. To discuss the best diagnostic standard.2. MATERIALS AND METHODS21 cases confirmed pathologically as breast tumor were enrolled in this study, with total 31 lesions, in which 19 were malignant and 12 were benign. The imaging series included axial SE T1WI, axial fat-saturated FSE T2WI, sagital fat-saturated FSE T2WI and axial dynamically enhanced 3D-FSPGR. The image subtraction and time-signal intensity curve were achieved on work station. Through statistical analysis, three diagnostic criteria for benign and malignant tumors were made according to morphology, type of dynamic enhancement curve and the combination of the both respectively. ROC analytic method was used to evaluate those criteria.3. RESULTSThere was significant difference between benign and malignant tumors on morphology (including shape, margin and enhancement homogeneity). Benign tumors usually had regular shape, clear margin and homogeneous enhancement, while malignant tumors were commonly irregular or spiculated shaped with poorly defined margin and heterogeneous or peripheral rim-like enhancement.There was also significant difference between benign and malignant tumors on the values of E1, E2, E3 , Tmax and the type of time-signal intensitycurve. There was no significant difference between them on the value of E4 and the later. ROC analysis showed sub curve area of E1 was biggest (Az=0. 864), and became smaller and smaller from E2, E3, E4 to E5 (Az=0. 781, 0.706, 0.658, 0.594). I -type time-signal intensity curve suggested a benign tumor, Ill-type curve represented a malignant one, while in the tumors with II-type curve , 83.3% were malignant and 16.7% were benign.Four misdiagnosis using the criteria of morphology or dynamic enhancement curve type alone were corrected by Using the criteria of combined and dynamic enhancement curve. ROC analysis showed the sub curve area of the combined standard was biggest, followed by the morphologic one and curve type one in sequence, with the Az values 0. 967, 0. 963 and 0. 908, respectively.4. CONCLUTIONContrast enhanced MRI has very high sensitivity in detection of breast tumor. Morphologic features and kinetic characteristics after dynamic enhancement can help differentiate benign and malignant tumor. The combination of morphology and dynamic enhancement curve can improve the accuracy of breast MRI...