Evaluation of the improvement of image quality in dual-energy contrast-enhanced digital mammography.

Abstract

Contrast-enhanced digital mammography (CEDM) has been employed as a complementary imaging examination to digital mammography, especially in more complex cases such as dense breasts, due to its enhanced sensitivity and specificity. This examination encompasses two main techniques: dual-energy and temporal. Both utilize X-ray spectra with higher energies than those in digital mammography and involve a double exposure of the breast, necessitating justification for its usage. In this study, the Monte Carlo method with the PENELOPE code (v. 2018) + PenEasy (v. 2020) was employed to simulate radiation transport and image acquisition for digital mammography and CEDM in the dual-energy technique. The goal was to compare and quantify differences in the image qualities of these modalities. Breast simulations considered phantoms with a homogeneous distribution of glandular and adipose tissue, using anthropomorphic voxelized breast models generated computationally through the VICTRE platform. Both models included lesions within the breast, composed of glandular tissue and iodinated contrast with varying concentrations for simulating digital mammography and CEDM, respectively. Quantitative image analyses were conducted based on signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Results demonstrated a significant improvement of up to 10 times for homogeneous breasts and 4.5 times for the anthropomorphic model in image quality with CEDM, justifying its usage. The degree of lesion enhancement depends on breast glandularity and the concentration of the iodinated contrast solution used.