The objective of this study was to quantify the hemodynamic parameters using first pass analysis of T1-perfusion magnetic resonance imaging (MRI) data of human breast and to compare these parameters with the existing tracer kinetic parameters, semi-quantitative and qualitative T1-perfusion analysis in terms of lesion characterization.
Materials and methods
MRI of the breast was performed in 50 women (mean age, 44±11 [SD] years; range: 26–75) years with a total of 15 benign and 35 malignant breast lesions. After pre-processing, T1-perfusion MRI data was analyzed using qualitative approach by two radiologists (visual inspection of the kinetic curve into types I, II or III), semi-quantitative (characterization of kinetic curve types using empirical parameters), generalized-tracer-kinetic-model (tracer kinetic parameters) and first pass analysis (hemodynamic-parameters). Chi-squared test, t-test, one-way analysis-of-variance (ANOVA) using Bonferroni post-hoc test and receiver-operating-characteristic (ROC) curve were used for statistical analysis.
All quantitative parameters except leakage volume (Ve), qualitative (type-I and III) and semi-quantitative curves (type-I and III) provided significant differences (P<0.05) between benign and malignant lesions. Kinetic parameters, particularly volume transfer coefficient (Ktrans) provided a significant difference (P<0.05) between all grades except grade-II vs III. The hemodynamic parameter (relative-leakage-corrected-breast-blood-volume [rBBVcorr) provided a statistically significant difference (P<0.05) between all grades. It also provided highest sensitivity and specificity among all parameters in differentiation between different grades of malignant breast lesions.
Quantitative parameters, particularly rBBVcorr and Ktrans provided similar sensitivity and specificity in differentiating benign from malignant breast lesions for this cohort. Moreover, rBBVcorr provided better differentiation between different grades of malignant breast lesions among all the parameters.Le texte complet de cet article est disponible en PDF.
Keywords : Magnetic resonance imaging (MRI), Perfusion MR imaging, Breast tumor, Lesion grading, Quantitative and qualitative analysis