Fourier Multiscale Analysis of Bilateral Breast Asymmetry Associated with Breast Cancer

A substantial body of evidence demonstrates that mammographic measurements are associated with breast cancer. In this matched case-control study, mammograms acquired approximately at the time of diagnosis were analyzed. The objective is to investigate bilateral breast asymmetry as an image measure that may reflect a continuum of image characteristics spanning breast cancer risk through disease-related alterations. Specifically, contralateral breast images were compared with measures derived in the Fourier domain (FD); this technique summarizes power in concentric radial bands that cover the Fourier plane. Equivalently, this approach can be described as a multiscale characterization of the image. The summarized power difference between respective contralateral bands produces a bilateral asymmetry measure (ASM). Full field digital mammography (FFDM) and synthetic two-dimensional images from digital breast tomosynthesis (DBT) were investigated for women that had both types of mammograms acquired at the same time. Conditional logistic regression modeling was used to estimate odds ratios (ORs) for the ASM in per standard deviation increment and the area under the receiver operating characteristic curves (Azs) with 95% confidence intervals; models included body mass index and ethnicity as controlling variables. Raw unprocessed FFDM mammograms (70 μm pixel spacing) produced significant findings: OR = 1.90 (1.58, 2.29) and Az = 0.72 (0.67, 0.76). Associations were significant but attenuated for both clinical FFDM (70 μm pixel spacing) and DBT mammograms (pixel spacing ranging from 90 -110μm): OR = 1.31 (1.11, 1.54) and Az = 0.63 (0.58, 0.67); and OR = 1.48 (1.25, 1.76) and Az = 0.65 (0.60, 0.70), respectively. Clinical FFDM and DBT images were inferior to raw FFDM images for capturing bilateral breast asymmetry. DBT images have lower spatial resolution than clinical FFDM images but produced stronger associations. These findings provide strong support for investigating this bilateral ASM in studies specifically designed to evaluate breast cancer risk prediction and early detection outside the matched case-control setting.