This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Breast, Mammography, Dosimetry, Dosimetric comparison
Authors:
C. Tromans, A. Chan, R. Highnam; Wellington/NZ
DOI:
10.1594/ecr2014/C-0360
Aims and objectives
Mammographic screening is associated with a substantial and significant reduction in breast cancer mortality [1].
However,
the breast is a highly radiation-sensitive organ,
and although low,
the risk of a radiation-induced cancer is concerning to many women,
and can adversely affect adherence to screening programs [2].
For routine screening,
it is therefore important that x-ray dose is kept as low as possible without compromising on image quality or the ability to detect cancers.
To keep patient doses at a minimum,
accurate calculations for monitoring and tracking dose are essential,
and methods for estimating dose that take into account individual patient characteristics are needed.
Most mammography manufacturers estimate a mean glandular dose (MGD) for each exposure taken and insert that value into a DICOM tag for the image.
However,
there are many different algorithms for estimating MGD and it is not clear which manufacturers use which algorithm.
Further,
all the algorithms need an estimate of glandularity in the breast and,
again,
it is not clear what assumptions each manufacturer makes.
Therefore,
given the wide range of glandularity that exists in the population,
radiation dose may be under- or over-estimated in the manufacturer-reported doses [3,
4].
Data from the large DMIST trial demonstrated that large variations in the average MGD reported by x-ray manufacturers exist [5].
For digital mammography,
mean dose per view ranged from 1.78- 2.50 mGy across several manufacturers’ systems.
Without a standardized method for estimating dose,
it is difficult to determine whether such variation is due to patient factors (e.g.
glandularity),
technologist factors (e.g.
degree of breast compression),
manufacturer factors (e.g.
varying dose algorithms and different detector technologies),
or a combination of these.
In this study,
the manufacturer-reported MGDs in the DICOM headers were compared to patient-specific MGDs generated using the woman’s specific glandularity and using the same MGD estimation algorithm.