Background Restriction spectrum imaging (RSI) is an advanced quantitative diffusion-weighted magnetic resonance imaging (DWI) technique to assess breast cancer. Purpose To investigate the ability of RSI to differentiate the benign and malignant breast lesions and the association with prognostic factors of breast cancer. Study Type Retrospective. Population Seventy women (mean age, 49.6 +/- 12.3 years) with 56 malignant and 19 benign breast lesions. Field Strength/Sequence 3-T; RSI-based DWI sequence with echo-planar imaging technique. Assessment The apparent diffusion coefficient (ADC) and RSI parameters (restricted diffusion f(1), hindered diffusion f(2), free diffusion f(3), and signal fractions f(1)f(2)) were calculated by two readers for the whole lesion volume and compared between the benign and malignant groups and the subgroups with different statuses of prognostic factors in breast cancer. Statistical Tests Mann-Whitney U test or Student's t-test was applied to compare the quantitative parameters between the different groups. Intraclass correlation coefficient (ICC) was used to assess readers' reproducibility. Binary logistic regression was used to combine parameters. Area under the curve (AUC) of receiver operating characteristic curve analysis was used to evaluate the diagnostic performance of parameters to distinguish benign from malignant breast lesions. A P-value Malignant breast lesions showed significantly lower ADC and f(3) values, and significantly higher f(1) and f(1)f(2) values than the benign lesions, with AUC of 0.951, 0.877, 0.868, and 0.860, respectively. When RSI-derived parameters and ADC were combined, the diagnostic performance was superior to either single parameter (AUC = 0.973). The f(3) value was significantly differed between estrogen receptor (ER)-positive and ER-negative tumors. The ADC, f(1), f(3), and f(1)f(2) values were significantly different progesterone receptor (PR)-positive and PR-negative status. Data Conclusion The RSI-derived parameters (f(1), f(3), and f(1)f(2)) may facilitate the differential diagnosis between benign and malignant breast lesions. Level of Evidence 4 Technical Efficacy Stage 2