This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Cardiac, Cardiovascular system, Radioprotection / Radiation dose, CT, CT-Angiography, CT-Quantitative, Diagnostic procedure, Arteriosclerosis, Calcifications / Calculi
Authors:
G. J. Pelgrim1, M. Meyer2, M. Vonder1, M. Greuter1, S. Huijsse1, P. M. Van Ooijen1, M. Oudkerk1, T. Henzler2, R. Vliegenthart1; 1Groningen/NL, 2Mannheim/DE
DOI:
10.1594/ecr2015/C-0622
Methods and materials
An anthropomorphic thoracic phantom with a cylindrical insert containing 100 small calcium spots was used.
(Figure 1) The HU density of the cylindrical phantom is comparable to unenhanced myocardium.
The 100 calcium spots are divided into four planes of 25 spots.
Spots varied in density (range 92-548 HA/cm3) and diameter (range 0.5-2 mm).
Physical reference volumes and masses were known.
The phantom was scanned on a third generation dual source CT scanner (Somatom Force,
Siemens,
Forchheim) at tube voltages of 70,
100 and 120 kV.
Tube current was selected using Caredose at reference mAs of 66.
Other scan parameters were: 192 x 0.6 mm acquired slice thickness,
with a gantry rotation time of 250 ms.
Reconstructed slice thickness was 3.0 mm/3.0 mm,
with a field of view of 250 mm.
All scans were obtained using the high-pitch spiral acquisition mode.
The QRM phantom was placed on the table in the middle of the gantry.
The phantom was not moved during the experiment and acquisition starting point was the same for every acquisition.
Reconstruction methods included filtered back projection and five grades of IR (Advanced Modeled Iterative Reconstruction (ADMIRE),
Siemens,
Forchheim),
a total of six reconstructions.
Calcium spot detectability was determined using a Matlab script.
A spot was ‘detected’ if two or more neighbouring voxels had a CT density of more than 130 HU in a 3D plane.
Agatston,
volume and mass scores were calculated automatically using the same dedicated Matlab script.
Reference standards used were EBT Agatston score (130kV,
600mAs,
3mm slice thickness),
available from a previous experiment and physical volume and mass.
An article by Groen et al determined the Agatston score for EBT in the same phantom,
using the same setup.[4]
Total reference volume and mass were determined for each acquisition separately,
based on the detected spots,
to correct for differences in detection.
Difference between the measured volume and mass score and physical values were normalized by dividing by the physical volume or mass.
Absolute percentile differences were then used to calculate the mean differences per calcium spot.
Using this method,
differences in detection rate,
volume and density were corrected for.
A value of zero means that there is no difference between measured score and physical value,
and a value of one means that a score that is 100% bigger or smaller compared to the physical value.