Changes in CT number of high atomic number materials with
field of view when using an extended CT number to electron
density curve and a metal artifact reduction reconstruction
algorithm
Poster No.:
R-0094
Congress:
2014 CSM
Type:
Scientific Exhibit
Authors:
V. Nelson, A. Gray; CAMPBELLTOWN/AU
Keywords:
Radiation physics, CT, Radiation therapy / Oncology,
Radiotherapy techniques
DOI:
10.1594/ranzcr2014/R-0094
Any information contained in this pdf file is automatically generated from digital material
submitted to EPOS by third parties in the form of scientific presentations. References
to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not
in any way constitute or imply RANZCR/AIR/ACPSEM's endorsement, sponsorship
or recommendation of the third party, information, product or service. RANZCR/AIR/
ACPSEM is not responsible for the content of these pages and does not make any
representations regarding the content or accuracy of material in this file.
As per copyright regulations, any unauthorised use of the material or parts thereof as
well as commercial reproduction or multiple distribution by any traditional or electronically
based reproduction/publication method ist strictly prohibited.
You agree to defend, indemnify, and hold RANZCR/AIR/ACPSEM harmless from and
against any and all claims, damages, costs, and expenses, including attorneys' fees,
arising from or related to your use of these pages.
Please note: Links to movies, .ppt slideshows, .doc documents and any other multimedia
files are not available in the pdf version of presentations.
Page 1 of 13
Aim
In external beam radiation therapy treatment planning CT numbers are used to perform
tissue heterogeneity corrections. The presence of metallic hip prostheses in patients with
pelvic malignancies produces artifacts which alters the CT numbers. Philips Healthcare
provides an image reconstruction algorithm, O-MAR, which reduces artifacts caused by
metal objects in CT images (Fig.1). In addition to this a 16-bit reconstruction is also
provided which extends the range of CT numbers. The purpose of this study was to
investigate the impact of bit depth reconstruction and the OMAR correction on the CT
number of high atomic number (Z) materials.
Images for this section:
Page 2 of 13
Fig. 1
Page 3 of 13
Methods and materials
CT scans were performed with a Philips Brilliance 16 slice CT scanner on a tissue
characterisation phantom (CT number to electron density (ED) phantom) (Fig.2). The
center of the phantom was aligned with the center of the bore. All scans used the
same scan range to ensure that the scans and their reconstructions would geometrically
coincide in order to facilitate data analysis. Three sets of scans were made using our
clinical scanning parameters (120kVp, 250 mAs/slice, 2 mm slice width, standard filter 'C'
for filtered back projection). In the first scan, the phantom was scanned in its homogenous
configuration and the images were reconstructed without O-MAR (12-bit depth) to obtain
the metal-free baseline CT number values. The second set of scans was performed with
3
a titanium (#=4.59 g/cm ) rod inserted in the phantom and images were reconstructed
without and with O-MAR (Figs. 3(a) and 3(b)) using 12 and 16-bit depths. The third
set of scans was performed with a steel insert (Figs. 3(c) and 3(d)) and the image
reconstructions were performed as for the second scan. The scans were completed
using two clinical CT scanning protocols, brain and pelvis. CT number to ED curves
derived from this data were compared for a change in the CT number due to the O-MAR
reconstruction and the 16-bit depth reconstruction. A planning study was also performed
on a commercial radiotherapy treatment planning system using a water equivalent slab
2
phantom at 100 cm SSD with a 10x10 cm 6MV photon beam for a fixed MU delivery. A
1 cm diameter cylindrical region of interest (ROI) was created in the phantom at a depth
of 5 cm. Doses were computed at five points beyond the ROI (Fig 4). In the first plan the
ROI was assigned a density of 4.59, with a CT to ED file generated from 12-bit images
with a titanium insert. For the planning system used in the study, 4.59 was the maximum
density that could be assigned when the CT-ED curve based on 12-bit images was used.
This plan was then recalculated using a CT to ED file generated from 16-bit images with
titanium and steel inserts. In the second plan the ROI was assigned a density of 8.1 to
represent steel and an extended CT to ED file, generated from 16-bit images with titanium
and steel inserts, was used. The plan with the maximum allowable density of 4.59 and
the 12-bit image CT-ED curve was used for comparison with this plan.
Images for this section:
Page 4 of 13
Fig. 2
Page 5 of 13
Fig. 3
Page 6 of 13
Fig. 4
Page 7 of 13
Results
The mean CT numbers from the 16-bit with and without O-MAR reconstructions were
in good agreement with the baseline values from the 12-bit without O-MAR scans, with
tissue equivalent materials being within 20 for most cases (Tables 1 and 2). For 12-bit
images, the CT numbers for titanium and stainless steel saturated at 3052 Hounsfield
units (HU) for O-MAR uncorrected images and 3049 for O-MAR corrected images. For 16bit depth images, the mean CT numbers of titanium and steel were higher than for the 12bit images, but the difference between the values for the with and without O-MAR scans
were within one standard deviation of each other. The CT to ED curves produced from the
12-bit and 16-bit reconstruction were similar below electron densities of approximately
1.7, but differed significantly beyond an electron density of 2.0 (Figure 5). In the planning
study it was found that the doses at all the measurement points beyond the high Z object
location were equivalent when the ROI density was assigned to 4.59 for the 12 and 16bit CT-ED curves. However, when the ROI density was assigned to 8.1 the doses were
lower when the extended CT to ED file was used when compared with doses calculated
using the CT to ED file generated with 12-bit image reconstruction data (Table 3). This
difference was more than 7% at 2 and 5 cm beyond the calculation point. However, there
was no significant difference in doses at surface point and at 1 cm in front of ROI.
Images for this section:
Page 8 of 13
Table 1
Page 9 of 13
Table 2
Page 10 of 13
Fig. 5
Table 3
Page 11 of 13
Conclusion
16-bit O-MAR-corrected images were found to result in an increased range of maximum
CT number allowable and as a result may provide a more accurate estimation of doses
in tissues surrounding high Z materials. Provided the planning system is able to accept
this image type, 16-bit images are recommended. Applying the O-MAR correction did not
have a large impact on the CT numbers so the CT to ED curve for the 16 bit images can
be applied for images with or without the correction applied.
Personal information
Vinod Nelson
Medical Physicist
MacArthur Cancer Therapy Centre
Campbelltown NSW 2560
Alison Gray
Medical Physics Specialist
Liverpool & MacArthur Cancer Therapy Centre
Campbelltown NSW 2560
References
1.
2.
3.
Glide-Hurst C, Chen D., Zhong H and Chetty I J. Changes realized from
extended bit-depth and metal artifact reduction in CT. Med. Phys. 2013
Volume 40 (No 6): p061711.
Spadea M F, Verburg J, Baroni G and Seco J. Dosimetric assessment of
a novel metal artefact reduction method in CT images. J/ Appl/ Clin/ Med.
Phys. 2013 Volume 14 (No 1): p 299
Philips Health Care Metal Artifact Reduction for Orthopedic Implants
(O-MAR). USA. Philips CT Clinical Science. Available from: http://
clinical.netforum.healthcare.philips.com/us_en/Explore/White-Papers/CT/
Metal-Artifact-Reduction-for-Orthopedic-Implants-%28O-MAR%29.
Page 12 of 13
Page 13 of 13