Abstract: Objective: To evaluate the diagnostic value of breast magnetic resonance imaging (MRI) in differentiating benign and malignant lesions classified as Breast Imaging Reporting and Data Systems (BI-RADS) category 4 on conventional X-ray mammography. Methods: A total of 110 patients with breast lesions classified as BI-RADS category 4 on conventional mammography, who were examined at our hospital between September 2022 and January 2025, were retrospectively enrolled. MRI morphological features and quantitative parameters were compared between benign and malignant lesions. Results: Among the 110 patients, 74 were malignant and 36 were benign based on pathological examination. Malignant lesions more frequently exhibited irregular shape, non-smooth margins, and a plateau-type time signal intensity curve (TIC), with proportions of 83.78%, 86.49%, and 64.86%, respectively; these values were significantly higher than those observed in benign lesions. The MRI quantitative parameter volume transport constant (Ktrans) was significantly higher in malignant lesions than in benign lesions (0.61 ± 0.20 min⁻¹ vs. benign lesions, P < 0.05), whereas the apparent diffusion coefficient (ADC) value was significantly lower in malignant lesions (1.11 ± 0.16) × 10⁻³mm²/s, P < 0.05). Logistic regression analysis showed that lesion morphology, lesion margin, TIC curve type, Ktrans, and ADC values are significant factors associated with malignancy. ROC curve analysis showed that the area under the curve of the model for diagnosing malignant lesions was 0.854 (95% CI: 0.805–0.902, P < 0.05); the sensitivity and specificity were 75.20% and 85.90%, respectively. Conclusion: Breast MRI demonstrates good diagnostic performance in differentiating benign and malignant lesions as BI-RADS category 4 on conventional X-ray mammography and may provide complementary information for clinical decision-making.