Research Article

Similarity of Fibroglandular Breast Tissue Content Measured from Magnetic Resonance and Mammographic Images and by a Mathematical Algorithm

Figure 2

(a)–(d) Scatter plot matrix including Pearson and regression line for pairwise correlation analyses between %-breast density (a), fibroglandular tissue volumes (b), adipose tissue volumes (c), and total breast tissue volumes (d) measured by five different methods. Diagonal boxes show histograms for each variable. Mamo_HSM (1st row and 1st column), histogram segmentation method using mammograms; Mamo_FFDM (2nd row and 2nd column), mammograms from full field digital mammography; Mamo_MATH (3rd row and 3rd column), mathematical algorithm for computing breast tissue content using mammograms; MRI_3DGRE (4th row and 4th column), 3-dimensional gradient-echo pulse sequence using MRI images; MRI_STIR (5th row and 5th column), short tau inversion recovery pulse sequence using MRI images. Units of measures for -axis and -axis are score (mean = 0, standard = 1) for (a)–(d). All data used for each pairwise correlation analysis are included within the graph ruler space. The bin width within each histogram is equally distributed within the column -axis scale and frequency in -axis is not labeled but represents relative distribution. For mean values, consult Figure 1.
961679.fig.002a
(a) Correlation matrix scatter plots, Pearson and histograms for %-breast density measures
961679.fig.002b
(b) Correlation matrix scatter plots, Pearson , and histograms for total breast volume measures
961679.fig.002c
(c) Correlation matrix scatter plots, Pearson , and histograms for glandular breast volume measures
961679.fig.002d
(d) Correlation matrix scatter plots, Pearson , and histograms for fat breast volume measures