Multiparametric MR-based feature fusion radiomics combined with ADC maps-based tumor proliferative burden in distinguishing TNBC vs. non-TNBC.

Objective： To investigate the incremental value of quantitative stratified ADC-defined tumor habitats for differentiating triple negative breast cancer (TNBC) from non-TNBC on multiparametric MRI (mpMRI)-based feature-fusion radiomics (R FF ) model.&#xD;Approach： 466 breast cancer patients (54 TNBC, 412 non-TNBC) who underwent routine breast MRIs in our hospital were retrospectively analyzed. Radiomics features were extracted from whole tumor on T2WI, DWI, ADC maps and the 2 nd phase of dynamic contrast-enhanced MRI. Four models including the R FF model (fused features from all MRI sequences), R ADC model (ADC radiomics feature), Stratified ADC model (tumor habitas defined on stratified ADC parameters) and combinational R FF -Stratified ADC model were constructed to distinguish TNBC vs. non-TNBC. All cases were randomly divided into a training (n=337) and test set (n=129). The four competing models were validated using the area under the curve (AUC), sensitivity, specificity and accuracy. &#xD;Main results： Both the R FF and Stratified ADC models demonstrated good performance in distinguishing TNBC from non-TNBC, with best AUCs of 0.818 and 0.773 in the training and test sets. Stratified ADC model revealed significant different tumor habitats (necrosis/cysts habitat, chaotic habitat or proliferative tumor core) between TNBC and non-TNBC with its top three discriminative parameters (p<0.05). The integrated R FF -Stratified ADC model demonstrated superior accuracy over the other three models, with higher AUCs of 0.832 and 0.784 in the training and test set, respectively (p<0.05). &#xD;Significance： The R FF -Stratified ADC model through integrating various tumor habitats' information from whole-tumor ADC maps-based Stratified ADC model and radiomics information from mpMRI-based R FF model, exhibits tremendous promise for identifying TNBC.