AUTOMATED-ANALYSIS OF MAMMOGRAPHIC DENSITIES

Information derived from mammographic parenchymal patterns provides on e of the strongest indicators of the risk of developing breast cancer. To address several limitations of subjective classification of mammog raphic parenchyma into coarse density categories, we have been investi gating more quantitative, objective methods of analysing the film-scre en mammogram. These include measures of the skewness of the image brig htness histogram, and of image texture characterized by the fractal di mension. Both measures were found to be strongly correlated with radio logists' subjective classifications of mammographic parenchyma (Spearm an correlation coefficients, R(s) = -0.88 and -0.76 for skewness and f ractal dimension measurements, respectively). Further, neither measure was strongly dependent on simulated changes in mammographic technique . Correlation with subjective classification of mammographic density w as better when both the skewness and fractal measures were used in com bination than when either was used alone. This suggests that each feat ure provides some independent information.