Contrast Optimization in Clinical Contrast-Enhanced Digital Mammography Images

摘要

CEDM is a radiological technique based on the use of digital mammography equipment and the injection of an iodinated contrast medium to enhance the visualization of tissues of interest. In previous works, our group has proposed a formalism for the use of dual-energy temporal CEDM, based on weighted subtraction of images, that has been applied with success to phantom data. This methodology requires the selection of ROIs by a radiologist, to determine the weight factors. In this work, we propose an alternative that improves the contrast in clinical images resulting from dual-energy temporal CEDM subtraction, while freeing the method from ambiguities due to the ROI selection by a radiologist. The new subtraction algorithm is based on the use of weight factors calculated pixel-by-pixel. The main result after evaluation of the methodology on images of 10 patients randomly chosen is a substantial improvement of the contrast (~5 times), reaching values that are similar to those obtained with single energy subtraction.