A novel approach to background subtraction in contrast-enhanced dual-energy digital mammography with commercially available mammography devices: Polychromaticity correction.

Abstract

Contrast-enhanced digital mammography is an image subtraction technique that is able to improve the detectability of lesions in dense breasts. One of the main sources of error, when the technique is performed by means of commercial mammography devices, is represented by the intrinsic polychromaticity of the x-ray beams. The aim of the work is to propose an iterative procedure, which only assumes the knowledge of a small set of universal quantities, to take into account the polychromaticity and correct the subtraction results accordingly. In order to verify the procedure, it has been applied to an analytical simulation of a target containing a contrast medium and to actual radiographs of a breast phantom containing cavities filled with a solution of the same medium. The reconstructed densities of contrast medium were compared, showing very good agreement between the theoretical predictions and the experimental results already after the first iteration. Furthermore, the convergence of the iterative procedure was studied, showing that only a small number of iterations is necessary to reach limiting values. The proposed procedure represents an efficient solution to the polychromaticity issue, qualifying therefore as a viable alternative to inverse-map functions.