Diffusion-weighted MR Imaging of Benign and Malignant Breast Lesions

Objective To determine the diagnostic potential of Diffusion-weighted Imaging to differentiate between benign and malignant breast lesions by comparapparent diffusion coefficient(ADC) among normal breast tissue, benign and malignbreast lesions. Methods Retrospective analysis of 177 pathologically confirmed brlesions examined using EPI-DWI and compare the mean ADC among normal brtissue, benign and malignant breast lesions. The cut-off ADC value for malignlesion was determined using receiver operating characteristic(ROC) curve analyThe diagnostic accuracy of DWI was compared in breast mass and non-mass lesioResults Mean ADC values for benign(112/177), malignant(65/177) breast lesions normal breast tissue(177) were(1.57±0.25)×10-3mm2/s,(1.11±0.32)×10-3ms and(1.72±0.22)×10-3mm2/s, respectively. With a cut-off value of 1.23×10-3ms for ADC in ROC analysis, 80.42% sensitivity and 92.13% specificity were achievAmong 177 lesions 144 were mass and 33 were non-mass. The mean ADC of malignmass lesions [(1.06±0.27)×10-3mm2/s] was statistically lower than that of benign o[(1.58±0.26)×10-3mm2/s]. The sensitivity and specificity of the ADC with a threshold[1.23×10-3mm2/s] was 85.17﹪ and 92.84﹪respectively. The mean ADC of malignnon-mass lesions [(1.25±0.40)×10-3mm2/s] was statistically lower than that of benones [(1.50±0.18)×10-3mm2/s]. The sensitivity and specificity of the ADC witthreshold of [1.36×10-3mm2/s] was 77.81﹪ and 80.35﹪respectively. Conclusion TADC is an effective parameter in distinguishing between malignant and benign brlesions. The cut-off values of ADCs are different between mass and non-mass lesioDWI of the breast can help differentiate benign and malignant breast lesions from norbreast tissue.