Sonographic Discrimination between Benign and Malignant Breast Lesions with Use of Disparity Processing

Abstract Rationale and Objectives The authors performed this study to evaluate whether a digital processing technique called disparity processing (DP) can improve the differentiation between benign and malignant breast masses at ultrasound (US) and, thus, reduce the number of benign lesions being sampled for biopsy. Materials and Methods During an US examination, a sonographer slightly varies the pressure of the probe on the breast surface. DP can be used to evaluate pairs of B scans that represent the different parts of this compression cycle. The apparent displacement of the tissue is measured with DP, and a new image, called a correlation map, is constructed. This new image illustrates the similarity between the speckle patterns around each point. The authors evaluated 25 solid lesions with DP. Results were compared with those from ( a ) histologic examination of specimens obtained with core or surgical biopsy or ( b ) cytologic examination of specimens obtained with fine-needle aspiration. Results All 25 lesions were correctly diagnosed with DP. There were no false-positive or false-negative findings. All malignancies demonstrated a relatively low-brightness halo around the lesion perimeter with evidence of discontinuity. All benign lesions were associated with relatively high-brightness, continuous halos. Conclusion The results suggest that DP can help differentiate benign from malignant masses.