Phantom-based study exploring the effects of different scatter correction approaches on the reconstructed images generated by contrast-enhanced stationary digital breast tomosynthesis

Abstract

Stationary digital breast tomosynthesis (sDBT) is an emerging technology in which the single rotating x-ray tube is replaced by a fixed array of multiple carbon nanotube-enabled sources, providing a higher spatial and temporal resolution. As such, sDBT offers a promising platform for contrast-enhanced (CE) imaging. However, given the minimal enhancement above background with standard operational tube settings and iodine dosing, CE breast imaging requires additional acquisition steps to isolate the iodine signal, using either temporal or dual energy subtraction (TS or DES) protocols. Also, correcting for factors that limit contrast is critical, and scatter and noise pose unique challenges during tomosynthesis. This phantom-based study of CE sDBT compared different postacquisition scatter correction approaches on the quality of the reconstructed image slices. Beam-pass collimation was used to sample scatter indirectly, from which an interpolated scatter map was obtained for each projection image. Scatter-corrected projections provided the information for reconstruction. Comparison between the application of different scatter maps demonstrated the significant effect that processing has on the contrast-to-noise ratio and feature detectability ([Formula: see text]) in the final displayed images and emphasized the critical importance of scatter correction during DES.

Variation in digital breast tomosynthesis image quality at differing heights above the detector

“…Puett et al used a similar methodology using a CIRS Model 020 BR3D phantom. 33 Images were imported into MatLab Ver. R2018b (The MathWorks Inc., Natick, Massachusetts) and opened for viewing.…”

“…The CIRS phantom is one of a few specifically designed phantoms for DBT 32 with materials designed to mimic adipose and gland tissues ‘swirled’ together. The CIRS phantom has previously been shown to be a valid phantom for this type of research where it was also been used where signal measurements have been taken 33 . Furthermore, a similar phantom to the CIRS phantom with ‘cloud‐like’ appearance rather than ‘swirls’, also manufactured by the CIRS Inc, has been used to measure the contrast‐to‐noise (CNR) ratio in DBT images 34 …”

“…The CIRS phantom has previously been shown to be a valid phantom for this type of research where it was also been used where signal measurements have been taken. 33 Furthermore, a similar phantom to the CIRS phantom with 'cloud-like' appearance rather than 'swirls', also manufactured by the CIRS Inc, has been used to measure the contrast-to-noise (CNR) ratio in DBT images. 34 An image of the phantom and individual slabs can be seen in Figure 1.…”

“…Image analysis was undertaken by measuring pixel values in the six contrast discs and background of the image. Puett et al used a similar methodology using a CIRS Model 020 BR3D phantom 33 . Images were imported into MatLab Ver.…”

Deep-learning convolutional neural network-based scatter correction for contrast enhanced digital breast tomosynthesis in both cranio-caudal and mediolateral-oblique views

“…14 Also, anti-scatter grid is usually not used in DBT image acquisitions due to oblique projection angles. [15][16][17] Therefore, post-acquisition image processing methods are usually employed to estimate scatter presented in DBT projection images and efficiently improve image quality without increasing patient dose.…”