Aims and objectives
Breast density (BD) is a strong cancer risk factor.
Studies have reported that the risk of developing breast cancer is four to six times larger in women with high BD than in those with low BD [1].
Thus,
it is important to objectively assess BD.
Many of these measures are based on 2D area methods,
which relate the BD to the degree of bright tissue in a mammogram.
However,
2D-area-based measurements suffer the limitation that the method does not aim to quantify the amount of...
Methods and materials
A physical 3D breast phantom (dimensions 122.4 mm x 208.3 mm x 33.2 mm) based on a real patient mammogram was designed to reproduce the pixel values throughout an anonymized unprocessed mammographic patient image in cranio-caudal view.
The patient breast had a typical distributed pattern of fibroglandular and fatty tissue without signs of lesions or calcifications and a compressed thickness of 32 mm.
The patient image was acquired with a Mammomat Inspiration® system (Siemens Healthineers,
Forchheim,
Germany) using the auto-segmentation mode of the automatic exposure...
Results
The VBD evaluated by Volpara™ on the patient’s image was 10.8%,
while the one based on the phantom image was 11.1%.
Figure 5(a) shows the 2D phantom density map obtained by Volpara™,
while in figure 5(b) the relative difference between the two attenuation maps (i.e.
Figure 2 and Figure 3) is presented together with the histogram of the values.
Assuming the skin thickness is equal to the mean skin thickness reported by Huang et al.
[10],
the mean relative difference (i.e.
[Ivolpara(x,y)- Iphantom(x,y)]/ Iphantom(x,y)) between...
Conclusion
In this study the accuracy of the Volpara™ software,
in estimating the volumetric breast density (VBD) and the 2D distribution of the fibroglandular tissue throughout the breast,
was investigated using a 3D printed single-material breast phantom.
An excellent agreement was found for the overall VBD between patient and phantom images,
with an absolute difference between patient VBD and phantom VBD of 0.3% (i.e.
11.1%-10.8%).
This confirmed that the phantom images could be used to represent patient images,
at least for Volpara density map analysis and...
Personal information
Christian Fedon currently works as a post-doctoral fellow at the Department of Radiology and Nuclear Medicine of Radboud University Medical Center - Nijmegen (The Netherlands) - within the Advanced X-ray Tomographic Imaging (AXTI) research group.
His research field is x-ray dosimetry,
focused on breast applications.
He is an active member of two joint workgroups of AAMP/EFOMP TG 282 and TG 383,
which aim to develop a new universal breast dosimetry methodology and a new breast phantom for quality control procedures.
He is a member of...
References
[1] Boyd NF,
Martin LJ,
Yaffe MJ,
Minkin S (2011) Mammographic density and breast cancer risk: current understanding and future prospects.
Breast Cancer Res 13: 223.
[2] Ng KM,
Yip CH,
Taib NAM (2012) Standardisation of clinical breast-density measurements.
Lancet.
Oncol.
13: 334-336.
[3] Shepherd JA,
Kerlikowske K,
Ma L,
Duewer F,
Fan B et al.
(2011) Volume of mammographic density and risk of breast cancer.
Cancer Epidemiol.
Biomarkers Prev 20: 1473-1482.
[4] Alonzo-Proulx O,
Jong R,
Yaffe M (2012) Volumetric breast density characteristics as...