MULTIRESOLUTION STATISTICAL-ANALYSIS OF HIGH-RESOLUTION DIGITAL MAMMOGRAMS

A multiresolution statistical method for identifying clinically normal tissue in digitized mammograms is used to construct an algorithm for separating normal regions from potentially abnormal regions; that is, small regions that may contain isolated calcifications, This is the in itial phase of the development of a general method for the automatic r ecognition of normal mammograms. The first step is to decompose the im age with a wavelet expansion that yields a sum of independent images, each containing different levels of image detail, When calcifications are present, there is strong empirical evidence that only some of the image components are necessary for the purpose of detecting a deviatio n from normal, The underlying statistic for each of the selected expan sion components can be modeled with a simple parametric probability di stribution function, This function serves as an instrument for the dev elopment of a statistical test that allows for the recognition of norm al tissue regions. The distribution function depends on only one param eter, and this parameter itself has an underlying statistical distribu tion, The values of this parameter define a summary statistic that can be used to set detection error rates, Once the summary statistic is d etermined, spatial filters that are matched to resolution are applied independently to each selected expansion image. Regions of the image t hat correlate with the normal statistical model are discarded and regi ons in disagreement (suspicious areas) are flagged, These results are combined to produce a detection output image consisting only of suspic ious areas, This type of detection output is amenable to further proce ssing that may ultimately lead to a fully automated algorithm for the identification of normal mammograms.