Xinming Liu PhD, Abstract Co-Author: Nothing to Disclose

Chao-Jen Lai PhD, Abstract Co-Author: Nothing to Disclose

Gary J. Whitman MD, Presenter: Nothing to Disclose

William Robert Geiser MS, Abstract Co-Author: Nothing to Disclose

Chris Chorng-Gang Shaw PhD, Abstract Co-Author: Investigator, Stellar Micro Devices

Scan equalization digital mammography (SEDM) is a technique to combine slot scanning and exposure equalization to improve low contrast performance in digital mammography. In this study, images obtained with anti-scatter grid at various exposure levels were used to simulate SEDM images and to investigate the improvement of image signal-to-noise ratios (SNRs).

An anthropomorphic breast phantom (Gammex 169 “Rachel”, Gammex RMI, Middleton, WI) was imaged at various exposure levels using a full-field digital mammography system (SenoGraphe 2000D, GEMS, Milwaukee, WI). The exposure equalization factors were computed based on a standard full-filed digital mammography (FFDM) image acquired in the AEC mode. The equalized image was simulated and constructed by superimposing a selected set of FFDM images acquired at 2, 1, 1/2, 1/4, 1/8, 1/16, and 1/32 times of the regular AEC timed technique (125 mAs). The equalized image was renormalized with the exposure equalization factors to result in similar appearance as with the regular FFDM image. Two sets of image were acquired to allow two identical but independently formed equalized images to be subtracted from each other to estimate the noise levels. Similarly, two identical but independently acquired regular FFDM images were subtracted to estimate the noise level, too. Corrections were made to remove the additional system noise added during image superimposition in forming the equalized image. SNRs over selected regions-of-interest (ROIs) were computed to compare the image quality.

Exposure equalization was found to result in substantially improved signal-to-noise ratios (SNRs) in dense tissue regions. The improvement was found to be less significant in less dense region. The SNR improvement was found to be 46% on average over selected ROIs. The SNRs over the selected ROIs were found to be within 10.3% from each other in the equalized image versus 29.4% in the regular FFDM image.

Exposure equalization was found to substantially improve image SNRs in dense tissue regions and result in more uniform image SNRs. This may lead to better low contrast performance in breast imaging tasks. Funding support CA104759, CA124585, CA13852, EB00117

Exposure equalization may potentially improve the detection and visualization of micro-calcifications (MCs) and soft tissue masses in digital mammography.

Liu, X, Lai, C, Whitman, G, Geiser, W, Shaw, C, Effects of Exposure Equalization on Image Signal-to-Noise Ratios in Digital Mammography: A Simulation Study with an Anthropomorphic Breast Phantom.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11012689.html