Purpose: To investigate whether adding diffusion-weighted imaging (DWI) of morphology and kinetic features using quantitative analysis improves the accuracy of discriminating malignant from benign breast masses. Materials and Methods: A total of 104 benign and 171 malignant lesions in 205 women were included. Thirteen texture and eleven morphology features were computed from each lesion using a semi-automated segmentation method. To increase prediction accuracy, a newly designed classification model, difference-weighted local hyperplane (DWLH), was used for statistical analysis of the combined effects of the features for predicting lesion type. The mean apparent diffusion coefficient (ADC) value for each lesion was calculated. Diagnostic performances of the model with and without DWI for the same lesion set were evaluated independently using receiver operating characteristics analysis. Results: Malignant lesions had lower mean ADCs than benign lesions. Combined morphological and kinetic features achieved a diagnostic accuracy of 0.93. Adding an ADC threshold of 1.37 × 103 mm2/s with a diagnostic accuracy of 0.87, increased the overall accuracy to 0.94. A multivariate model combining ADC values with six morphological and kinetic parameters most significantly discriminated malignant from benign lesions. Conclusion: ADC values, combined with morphological and kinetic features, improve the discriminatory power of breast lesions.