Related Experiment Video
Updated: Mar 2, 2026
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Diagnostic value of susceptibility-weighted imaging in differentiating benign and malignant breast lesions
Fadime Güven1, Muhammed Halid Yener1, Muhammed Emre Işıktaş1
1Department of Radiology, Faculty of Medicine, Atatürk University, 25240 Erzurum, Turkiye.
Background:
Non-contrast breast MRI protocols are becoming increasingly important and gaining popularity as a safer and faster alternative to contrast-enhanced imaging. However, distinguishing benign from malignant breast lesions without the use of contrast agents can sometimes be difficult; therefore, exploring the diagnostic contribution of alternative additional sequences has become increasingly relevant.
Purpose:
The diagnostic performance of SWI sequences in differentiating benign from malignant breast lesions was evaluated, and their contribution was compared with ADC values.
Materials And Methods:
This retrospective single-center diagnostic accuracy study included 129 women who underwent breast MRI between 2022 and 2025 and were found to have histopathologically confirmed benign or malignant mass lesions. Parameters obtained from SWI sequences, including the Intralesional Susceptibility Score (ILSS) and rim morphology, were analyzed together with ADC values. Group comparisons were conducted using the Mann-Whitney U and chi-square tests. Diagnostic performance was assessed using ROC curve analysis and the DeLong test. ROC analysis and rim morphology assessments were reported separately for both readers. Interobserver agreement was evaluated using Cohen's κ coefficient.
Results:
The cohort comprised 129 women (mean age 47.1 ± 12.5 years). Malignancies had higher ILSS values than benign lesions (2.75 vs 1.18; p < 0.001). For both readers, malignancy correlated with irregular margins and an incomplete rim (p < 0.001). Although the AUC for the ILSS (0.86) was above ADC (0.78), the contrast was not significant (p = 0.159); using both measures together increased AUC to 0.90.
Conclusion:
SWI achieved a level of diagnostic accuracy comparable to ADC. When interpreted together, the two sequences enhanced overall diagnostic confidence. Moreover, analysis of rim morphology provided additional qualitative information that aided lesion characterization. These results suggest that SWI could function as a useful complementary sequence in non-contrast breast MRI.
