Purpose: To assess the potential of machine learning with multiparametric MRI (mpMRI) for the early prediction of pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) and of survival outcomes in breast cancer patients. Materials and Methods: This IRB-approved prospective study included 38 women (median age 46.5 years; range 25-70 years) with breast cancer who were scheduled for NAC and underwent mpMRI of the breast at 3T with DCE, DWI and T2-weighted imaging prior to and after two cycles of NAC. For each lesion, 23 features were extracted: qualitative T2-weighted and DCE-MRI features according to BI-RADS, quantitative pharmacokinetic DCE features (mean plasma flow, volume distribution, mean transit time) and DWI apparent diffusion coefficient (ADC) values. To apply machine learning to mpMRI, eight classifiers including linear support vector machine (SVM), linear discriminant analysis (LDA), logistic regression (LR), random forest (RF), stochastic gradient descent (SGD), decision tree, adaptive boosting (AdaBoost) and extreme gradient boosting (XGBoost) were employed to rank the features. Histopathologic Residual Cancer Burden (RCB) class (with RCB 0 being a pCR), recurrence-free survival (RFS) and disease-specific survival (DSS) were used as the standards of reference. Classification accuracy with area under the receiving operating characteristic curve (AUC) was assessed using all the

• The Tree scoring system shows high diagnostic accuracy in mass and non-mass lesions. • The Tree scoring system reduces inter-reader variability related to reader experience. • The Tree scoring system improves diagnostic accuracy in non-expert readers.

• Region of interest placement significantly influences apparent diffusion coefficient of breast tumours. • Minimum and mean apparent diffusion coefficient perform best and are reproducible. • 2D regions of interest perform best and provide rapid measurement times.

Objectives The aim of this study was to assess the potential of noncontrast magnetic resonance imaging (NC-MRI) with diffusion-weighted imaging (DWI) in characterization of breast lesions in comparison to dynamic contrast-enhanced MRI (DCE-MRI) at 3 T. Materials and Methods Consecutive patients with conventional imaging (mammography, ultrasound) BI-RADS 4/5 findings were included in this institutional review board–approved single-center study. All underwent 3 T breast MRI including readout-segmented DWI, DCE, and T2-weighted sequences. Final diagnosis was defined by histopathology or follow-up (>24 months). Two experienced radiologists (R1, R2) independently assigned lesion conspicuity (0 = minimal to 3 = excellent) and BI-RADS scores to NC-MRI (readout-segmented DWI including apparent diffusion coefficient maps) and DCE-MRI (DCE and T2-weighted). Receiver operating characteristics, κ statistics, and visual grading characteristics analysis were applied. Results Sixty-seven malignant and 56 benign lesions were identified in 113 patients (mean age, 54 ± 14 years). Areas under the receiver operating characteristics curves were similar: DCE-MRI: 0.901 (R1), 0.905 (R2); NC-MRI: 0.882 (R1), 0.854 (R2); P > 0.05, respectively. The κ agreement was 0.968 (DCE-MRI) and 0.893 (NC-MRI). Visual grading characteristics analysis revealed superior lesion conspicuity by DCE-MRI (0.661, P < 0.001). Conclusions Diagnostic performance and interreader agreement of both NC-MRI and DCE-MRI is high, indicating a potential use of NC-MRI as an alternative to DCE-MRI. However, inferior lesion conspicuity and lower interreader agreement of NC-MRI need to be considered.