ANALYSIS OF SPICULATION IN THE COMPUTERIZED CLASSIFICATION OF MAMMOGRAPHIC MASSES

Spiculation is a primary sign of malignancy for masses detected by mam mography. In this study, we developed a technique that analyzes patter ns and quantifies the degree of spiculation present. Our current appro ach involves (1) automatic lesion extraction using region growing and (2) feature extraction using radial edge-gradient analysis. Two spicul ation measures are obtained from an analysis of radial edge gradients. These measures are evaluated in four different neighborhoods about th e extracted mammographic mass. The performance of each of the two meas ures of spiculation was tested on a database of 95 mammographic masses using ROC analysis that evaluates their individual ability to determi ne the Likelihood of malignancy of a mass. The dependence of the perfo rmance of these measures on the choice of neighborhood was analyzed. W e have found that it is only necessary to accurately extract an approx imate outline of a mass lesion for the purposes of this analysis since the choice of a neighborhood that accommodates the thin spicules at t he margin allows for the assessment of margin spiculation with the rad ial edge-gradient analysis technique. The two measures performed at th eir highest level when the surrounding periphery of the extracted regi on is used for feature extraction, yielding A(z) values of 0.83 and 0. 85, respectively, for the determination of malignancy. These are simil ar to that achieved when a radiologist's ratings of spiculation (A(z) = 0.85) are used alone. The maximum value of one of the two spiculatio n measures (FWHM) from the four neighborhoods yielded an A(z) of 0.88 in the classification of mammographic mass lesions.