Bloch–Siegert B(1)-Mapping Improves Accuracy and Precision of Longitudinal Relaxation Measurements in the Breast at 3 T

Variable flip angle (VFA) sequences are a popular method of calculating T(1) values, which are required in a quantitative analysis of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI). B(1) inhomogeneities are substantial in the breast at 3 T, and these errors negatively impact the accuracy of the VFA approach, thus leading to large errors in the DCE-MRI parameters that could limit clinical adoption of the technique. This study evaluated the ability of Bloch–Siegert B(1) mapping to improve the accuracy and precision of VFA-derived T(1) measurements in the breast. Test–retest MRI sessions were performed on 16 women with no history of breast disease. T(1) was calculated using the VFA sequence, and B(1) field variations were measured using the Bloch–Siegert methodology. As a gold standard, inversion recovery (IR) measurements of T(1) were performed. Fibroglandular tissue and adipose tissue from each breast were segmented using the IR images, and the mean T(1) was calculated for each tissue. Accuracy was evaluated by percent error (%err). Reproducibility was assessed via the 95% confidence interval (CI) of the mean difference and repeatability coefficient (r). After B(1) correction, %err significantly (P < .001) decreased from 17% to 8.6%, and the 95% CI and r decreased from ±94 to ±38 milliseconds and from 276 to 111 milliseconds, respectively. Similar accuracy and reproducibility results were observed in the adipose tissue of the right breast and in both tissues of the left breast. Our data show that Bloch–Siegert B(1) mapping improves accuracy and precision of VFA-derived T(1) measurements in the breast.