Histogram Analysis of Multiplexed Sensitivity Encoding in the Differentiation of Benign and Malignant Breast Lesions as Compared to BI-RADS

Motivation: To provide more accurate ADC values for clinical practice to improve the diagnostic specificity and avoid unnecessary biopsy. Goal(s): To assess the diagnostic value of histogram parameters based on multiplexed sensitivity encoding (MUSE) in benign and malignant breast lesions. Approach: Breast lesions were outlined and ADC histogram parameters were obtained with the Firevoxel software, then the differences between the benign and malignant groups were compared. Results: The ADC histogram parameters were statistically significant different between benign and malignant lesions, among which ADC_5% had the better differential diagnosis value. The combined model (ADC_5%, ADC_MUSE, BI-RADS) had the highest diagnostic performance. Impact: ADC histogram analysis is an emerging image analysis technology that can analyze the entire tumor tissue component, which has been widely applied for many systems and has shown good diagnostic performance.