Abstract
The B-mode image has become a popular diagnostic tool in breast ultrasoundfor describing the properties of breast tumors in morphology and texture. Inorder to better characterize scatterers in breast tumors, the Nakagami imagebased on the statistical distribution of raw ultrasound data has also beensuccessfully proposed. However, since morphological analysis, texture analysis,and Nakagami imaging actually supply information on different physicalcharacteristics of breast tumors, a functional complementation by combining theabove methods may provide more clues for classifying breast tumors. For thisreason, this study investigated a novel multiple-parameter-based analysis methodfor evaluating breast tumors, in which the tumor morphology, the scattererechogenicity, and the scatterer arrangements and concentrations are describedusing the B-mode, texture, and Nakagami images, respectively. To verify thevalidity of the concept, raw data were obtained from 100 clinical cases. Allpatients were examined by an experienced radiologist, and the results wereconfirmed by surgery and pathological examinations or biopsy. The contours ofthe tumor were described manually by a breast surgeon familiar with breastultrasound interpretation. Three morphological parameters, three texturefeatures, and the Nakagami parameter of benignancy and malignancy wereextracted. Fuzzy c-means clustering was applied to identify a tumor as benign ormalignant based on combining the parameters. The results indicated that therewould be a trade-off between sensitivity and specificity when combining the samephysical characteristics. However, the combination of the standard deviation ofthe shortest distance (morphology), the variance (texture), and the Nakagamiparameter concurrently allows both the sensitivity and specificity to exceed85%, making the performance to diagnose breast tumors better. © 2010 IEEE.