COMPUTERIZED CHARACTERIZATION OF MASSES ON MAMMOGRAMS - THE RUBBER BAND STRAIGHTENING TRANSFORM AND TEXTURE ANALYSIS

A new rubber band straightening transform (RBST) is introduced for cha racterization of mammographic masses as malignant or benign. The RBST transforms a band of pixels surrounding a segmented mass onto the Cart esian plane (the RBST image). The border of a mammographic mass appear s approximately as a horizontal line, and possible spiculations resemb le vertical lines in the RBST image. In this study, the effectiveness of a set of directional texture features extracted from the RBST image s was compared to the effectiveness of the same features extracted fro m the images before the RBST. A database of 168 mammograms containing biopsy-proven malignant and benign breast masses was digitized at a pi xel size of 100 mu m X 100 mu m. Regions of interest (ROIs) containing the biopsied mass were extracted from each mammogram by an experience d radiologist. A clustering algorithm was employed for automated segme ntation of each ROI into a mass object and background tissue. Texture features extracted from spatial gray-level dependence matrices and run -length statistics matrices were evaluated for three different regions and representations: (i) the entire ROI; (ii) a band of pixels surrou nding the segmented mass object in the ROI; and (iii) the RBST image. Linear discriminant analysis was used for classification, and receiver operating characteristic (ROC) analysis was used to evaluate the clas sification accuracy. Using the ROC curves as the performance measure, features extracted from the RBST images were found to be significantly more effective than those extracted from the original images. Feature s extracted from the RBST images yielded an area (A(z)) of 0.94 under the ROC curve for classification of mammographic masses as malignant a nd benign. (C) 1998 American Association of Physicists in Medicine. [S 0094-2405(98)00904-3].