A comparison of imaging methods in the evaluation of breast pathologies

Abstract

Evaluating the performance of magnetic resonance imaging (MRI) and comparison with mammography, ultrasound (US) and histopathology results in cases of women with suspicious breast lesions. Forty nine cases on which histopathology was performed included in the study. All cases were applied mammography and US, and then MRI. Biopsy or post-operational results have been generated. Among the 49 cases, 27 (55%) benign and 22 (45%) were malign lesions. Using mammography, of the 49 cases, 24 (49%) were deemed to be malign. True positive cases were 20 (41%). In ultrasound, 25 (51%) were malign, 24 (49%) were benign. True positive cases were 21 (43%). Lesions were detected in all of the 49 cases using MRI. All of the 22 (45%) malign cases were diagnosed as malign lesion. Sensitivities of mammography, US and MRI in detecting lesions were 83%, 95% and 100%, and specificities were 85%, 85% and 92% respectively. In MRI, all cases were applied dynamic contrast sequences, for the cases with lesions detected, time signal intensity (SI) curves were drawn. 23 cases were detected as Type 1 (47%), 2 cases were Type 2 (4%), 24 cases were Type 3 (49%) SI curve. According to SI curves sensitivity in detecting malignities was 95% and specificity was 81%. MRI has been found superior to mammography and US in detecting masses especially with its characteristics of higher spatial resolution, less binding of dynamic properties evaluation on user. There may be a decrease of unnecessary interventional operations in benignmalign detection of breast lesions using dynamic breast MRI.