Segmentation of breast ultrasound lesion boundary using texture-based multi-resolution method

Computer-aided characterization of a breast ultrasound lesion involves two steps: first, extracting features from the lesion whose boundary is pre-defined on the images, and then converting the features into mathematical models. Most methods assume that the boundaries of the lesions are pre-selected or outlined by sonographers or operators, because automated delineation of lesion boundary is not trivial and is a challenging task. The purpose of this study was to develop and evaluate an automated lesion boundary segmentation method that is based on texture-based, multi-resolution image analysis. One hundred ninety-seven breast ultrasound images containing solid breast lesions from 172 women (age 24-89 years, mean 38 years) were studied. Fifty-five of the 197 images were from 40 women with malignant lesions, and the remaining 142 were from 132 patients with benign lesions. Each breast lesion was identified by an operator who placed a rectangular region of interest (ROI) to widely encompass the lesion. The resolution of the image was compressed, at variable ratios depending on the ROI size, to reduce noise. Texture momentum was computed. A binary image was generated from the texture and pixel intensity parameters. Initial seed boundary was segmented from the binary image and then expanded to the original resolution using the boundary pixel intensity gradient information. The boundary of each breast lesion was delineated by a breast-imaging radiologist who was blinded to the computer-detected lesion boundary. The 'area match ratio' between the manually drawn boundaries and the automatically detected boundaries was computed. This ratio is equal to or less than unity (unity indicates that the areas match exactly). Overall, good agreement was seen between the multi-resolution segmentation method and the radiologist's manual delineation. The mean area match ratio was 0.87 ±0.02. We have developed a multi-resolution, texture-based method to segment the boundary of breast lesions. This method will facilitate full automation for the characterization of breast ultrasound lesions.