Table 1.
Patient and subtype demographics.
Fig 1.
Example of tissue segmentation performed of all cancer patients’ affected breast images.
At top left (a), a dynamic contrast-enhanced MRI exam at t3 is seen in the axial plane, illustrating one slice of the view used for contouring the breast and tumor. At top right (b), the result of breast segmentation is shown. At bottom left (c), the segmented tumor is highlighted in blue. Finally at bottom right (d), the parenchyma segmented at t1 is highlighted in pink. Breast subcompartment segmentation was performed in 3-dimensions.
Fig 2.
Summary of radiomic analysis performed in this study.
Clinical features were evaluated by a radiologist according to Breast Imaging Reporting and Data System directly from dynamic contrast-enhanced MRI (a). 3-Dimensional tumor (red) and parenchyma (light blue) compartments were segmented (b), from which volumetric breast density was immediately estimated (c). Enhancement maps were then generated (d), from which textural features of tissue compartments were extracted and defined as enhancement heterogeneity (e). Subsequently, two analyses were conducted using extracted features: supervised learning of breast cancer subtype was performed with a support vector machine classifier (f) and unsupervised learning of background parenchymal enhancement feature expression pattern was performed with k-means clustering (g).
Table 2.
Performance results of predictive modeling.
Table 3.
Optimal imaging features selected.
Imaging features most discriminative in prediction models, using tumor features (top 5 rows) or using both tumor and parenchyma features (bottom 5 rows). Features that survived the selection algorithm in the majority of cross-validation folds of the given task are shown with numbers indicating percentage of folds in which the feature was selected as a simplistic indicator of significance (see Supplemental Table, S1 Table, for further elaboration of feature values and their significance). Features are individually identified here at the bottom of the table, abbreviated by type and defined in the footnote.
Fig 3.
Box plots illustrating differences in distributions (quartiles as red boxes, grand mean indicated as spanning line) of the three most predictive quantitative features found in differentiation tasks: the lesion’s ‘mass size’ feature (a), parenchyma’s ‘skewness of Signal Enhancement Ratio’ feature (b), and parenchyma’s ‘standard deviation of rate in’ feature (c) compared between the triple-negative (TN) and non-TN groups.
p-values were calculated by Wilcoxon Mann-Whitney tests.
Fig 4.
Examples of ‘parenchyma rate in’ parameter (also Fig 3B) maps from a non-triple-negative (TN) patient (left) and a TN patient (right) illustrating the difference of a statistical texture feature between members of the two groups in image form.
Slices of the ‘parenchyma rate in’ parameter map void of tumor tissue are presented in the sagittal plane. It is evident the variation of this background parenchymal enhancement texture feature’s value is greater in TN cancers, where standard deviation is markedly higher at 352.9 as opposed to 133.8 in the non-TN patient.
Fig 5.
Unsupervised k-means clustering of breast cancer patients (n = 88) on the x-axis and quantitative background parenchymal enhancement (BPE) feature expression (n = 39) on y-axis (as z-scores, with scale at bottom left. std = standard deviation).
Correspondence of patient groups with similar radiomic expression patterns can be seen where the majority of triple-negative (TN) breast cancers have grouped together in the left cluster (9 of 11 TN in partition highlighted orange at top left) due to association of the BPE heterogeneity feature signatures. 1st order statistical texture features are highlighted as purple and similarly 2nd order statistical texture features are green at right indicating correspondence of feature groups with clustered expression patterns.