Design and validation of realistic breast models for use in multiple alternative forced choice virtual clinical trials

Abstract

A novel method has been developed for generating quasi-realistic voxel phantoms which simulate the compressed breast in mammography and digital breast tomosynthesis (DBT). The models are suitable for use in virtual clinical trials requiring realistic anatomy which use the multiple alternative forced choice (AFC) paradigm and patches from the complete breast image. The breast models are produced by extracting features of breast tissue components from digital breast tomosynthesis (DBT) clinical images including skin, adipose and fibro-glandular tissue, blood vessels and Cooper's ligaments. A range of different breast models can then be generated by combining these components. Visual realism was validated using a receiver operating characteristic (ROC) study of patches from simulated images calculated using the breast models and from real patient images. Quantitative analysis was undertaken using fractal dimension and power spectrum analysis. The average areas under the ROC curves for 2D and DBT images were .51±.06 and .54±.09 demonstrating that simulated and real images were statistically indistinguishable by expert breast readers (7 observers); errors represented as one standard error of the mean. The average fractal dimensions (2D, DBT) for real and simulated images were (2.72±.01, 2.75±.01) and (2.77±.03, 2.82±.04) respectively; errors represented as one standard error of the mean. Excellent agreement was found between power spectrum curves of real and simulated images, with average β values (2D, DBT) of (3.10±.17, 3.21±.11) and (3.01±.32, 3.19±.07) respectively; errors represented as one standard error of the mean. These results demonstrate that radiological images of these breast models realistically represent the complexity of real breast structures and can be used to simulate patches from mammograms and DBT images that are indistinguishable from patches from the corresponding real breast images. The method can generate about 500 radiological patches (~30mm x 30mm) per day for AFC experiments on a single workstation. This is the first study to quantitatively validate the realism of simulated radiological breast images using direct blinded comparison with real data via the ROC paradigm with expert breast readers. Page 1 of 16AUTHOR SUBMITTED MANUSCRIPT -PMB-104914. R2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 KeywordsVirtual clinical trials, breast model, breast phantom, synthetic breast, image simulation, 2D-mammography, digital breast tomosynthesis.

The threshold detectable mass diameter for 2D-mammography and digital breast tomosynthesis

“…Clinical breast readers were unable to distinguish simulated image patches produced using these phantoms from equivalent image patches from real mammograms and DBT images. Furthermore, the statistical properties of simulated images showed a strong match to those of real images [25].…”

“…For consistency with the methods followed by Hadjipanteli et al [18], we used six realistic mathematical breast phantoms validated for 4-AFC studies [25]. The breast phantoms were produced using a biological approach in which breast components were simulated using features extracted from clinical DBT images.…”

“…The aim of this work therefore was to use a simulation-based observer study to quantify the threshold detectable mass diameter for 2Dmammography, narrow and wide angle DBT. For this purpose we simulated the images of a voxel phantom [25] with inserted masses to create realistic simulated images. An advantage of this approach is that the same breast can be imaged using different acquisition methods, unlike clinical studies.…”

Design and validation of biologically inspired spiculated breast lesion models utilizing structural tissue distortion

“…A series of experiments was undertaken using our synthetic breast phantoms 6 to understand the ways in which tissues can appear to distort in the local vicinity of a mass, and so attempt replication of these observed distortions and twists. The approach taken was primarily aimed at producing a realistic appearance as seen in clinical images, and so may not necessarily be representative of actual distortions present in real breast tissue.…”

“…A(r The OPTIMAM image simulation toolbox 2 is a complete end-to-end VCT pipeline that can generate synthetic masses 3,4 or calcifications 5 and insert these into unprocessed clinical digital mammography images or hybrid models of breast tissue 6 . These are then processed using an image acquisition tool that can currently simulate either the Hologic Selenia 2 or the Siemens MAMMOMAT Inspiration 7 imaging systems in 2D and tomosynthesis operating modes.…”