Anatomical noise in contrast‐enhanced digital mammography. Part I. Single‐energy imaging

Abstract

The presence of an iodinated contrast agent in the breast produced small, but significant changes in the power law parameters of unprocessed CEDM images compared to the precontrast images. Image subtraction in SE CEDM significantly reduced anatomical noise compared to conventional DM, with a reduction in both α and β by about a factor of 2. The data presented here, and in Part II of this work, will be useful for modeling of CEDM backgrounds, for systems characterization and for lesion detectability experiments using models that account for anatomical noise.

Optimization of contrast‐enhanced breast imaging: Analysis using a cascaded linear system model

“…The methods for the experimental measurement of S B have been outlined in detail previously . The empirical values of S B used in the present work were analyzed using projection images of 4 cm of a BR3D structured noise phantom (Model 020; CIRS Inc., Norfolk, VA, USA).…”

“…Image subtraction is often used in CE imaging to remove background tissue structure, and the most common techniques are dual energy (DE) subtraction and temporal (TE) subtraction. Image subtraction as a matter of mathematical theory has been discussed in good detail previously . However, for the applications pertaining to the present work, it may be described and simplified as: where P denotes the log‐transformed projection images for the subtraction, mask ( M ), and contrast ( C ) projection views.…”

The effect of amorphous selenium detector thickness on dual‐energy digital breast imaging

“…Contrast enhanced (CE) x-ray breast imaging for planar techniques, such as digital mammography (DM) [1][2][3][4][5][6][7][8][9] and threedimensional (3D) techniques such as digital breast tomosynthesis (DBT) [10][11][12][13][14][15] has been the subject of intensive investigation. The conspicuity of large, malignant lesions may be enhanced through the injection of radio-opaque contrast agents (i.e., iodine).…”

Generation of 3D synthetic breast tissue

“…The standard error of the AUC for each observer was computed using the methods of Hanley and McNeil 10 . According to the previous literature 8 , structures in a mammogram can be described by a power law spectrum of form P(f)=A/f β , where β is the power-law exponent and A is the power spectrum magnitude. Figure 4 shows a comparison of the power spectra of simulated and real 2D image segments (averaged over 300 segments) computed using the method described in Hill et al 8 and Cockmartin et al 9 .…”

“…The power-law parameters were estimated by applying a linear fit to the log transformed data in the frequency range of 0.2-0.7mm -1 , as this range represents the variations due to breast anatomy 8,9 . The average β values for simulated images were 3.19±.07 (2D) and 3.21±.11 (DBT), and for real images were 3.01±.32 (2D) and 3.1±.17 (DBT); the errors are the standard errors of the mean β.…”