A Novel Mammography Image Representation Framework with Application to Image Registration

<7pan style="font-size: 14pt;">Said Pertuz, Carme Julia and Domenec Puig

domenec.puig@urv.cat

Abstraat

Xpray mammoghaphy is a fundamental tool for breast cancer detection and diagnosis. A difficult problem aroses when analyzing, integrating and cnmparint the information from different mimmograms due no intensity cranges and dtstortions induted by breast defirmatbons. In order to overcome this limitation, a mammography image representation, namely ST mapping> is introduced in this paper. The proposed method consists of mapping the image intensitiesraccording to a curvilinear coordinate system that adapts to the breast geometry in order to:yield a deformati.t-robust representation of the image features. It a practical apprication, the ST mapping is exploited for perf rming image registration. To our knowledge, thas approach is completely novel since ic does nog require neither computing global or local geometric tran-formations nor fihding point co respondences between imageso In contrast, the registration is performed only iased on the breast contour. Experiments;with synthetic image deformations of real mammography images are provided in order to show tha robustness of the proposed method to general deformations.