Objective: To describe imaging characteristics and evaluate the diagnostic value of multiparametric ultrasound for differentiating benign and malignant breast masses indicated for biopsy.
Methods: A cross-sectional diagnostic accuracy study included 84 patients with 105 breast masses examined using B-mode ultrasound, color Doppler, strain elastography, and shear-wave elastography between July 2024 and April 2026. Histopathology served as the reference standard. Logistic regression and receiver operating characteristic analysis were used to identify predictors and assess diagnostic performance.
Results: 76 masses were benign and 29 were malignant. Non-parallel orientation, indistinct/angular/spiculated margins, and SWE > 56 kPa independently predicted malignancy, with adjusted odds ratios of 6.196, 7.816, and 27.514, respectively. The AUCs of B-mode, Doppler, and elastography were 0.840, 0.628, and 0.929. The B-mode plus elastography model and the complete model achieved comparable AUCs of 0.961 and 0.963, respectively. The complete model yielded 93.1% sensitivity, 92.1% specificity, 97.2% negative predictive value, and 92.4% accuracy.
Conclusion: Combining B-mode ultrasound with elastography, particularly SWE, provides high diagnostic value for differentiating benign and malignant breast masses. Doppler provided supplementary information but did not substantially improve the AUC once B-mode ultrasound and elastography were incorporated into the model.