Transcript Evaluation of the Performance of the Fast Scanning Platform of an

Evaluation of the Performance of the Fast Scanning Platform of an OCT System
Malcolm
1Department
1
Heard ,
Miguel
1
Herrera ,
Geoffrey
1
Ibbott
of Radiation Physics, The University of Texas, M.D. Anderson Cancer Center, Houston, Texas
Introduction
Laser
Diffuser
Rotating
Mirror
Detector
Fresnel Lens
Fig. 1 Picture of the OCT system with the fast-scanning platform.
Materials and Methods
1.2
1
0.8
0.6
0.4
0.2
0
0
20 40 60 80 100 120 140 160 180 200 220
Scan Time (min)
x-ray CT Scans
180 degree mode
360 degree mode
50000
40000
30000
10000
0
0.4
0.2
0
0
20 40 60 80 100 120 140 160 180 200 220
Scan Time (min)
Fig. 3 Percent difference between the first scan and successive
scans acquired over a 3 hour period using the original configuration
of the OCT system.
% Standard Deviation
Percent Difference (%)
0.8
0.6
10
15
Fig. 8 Profiles through gels irradiated with half-blocked field and
scanned with the 180° and 360° modes of the fast scanning
platform.
20000
1
2
3
4
5
6
7
Fig. 6 Dose response of a polymer gel scanned with the 180° and
360° modes of the fast scanning platform. The error bars represent
one standard deviation.
1
5
At each dose level, the OD values measured
with the 180° mode were within one standard
deviation of the OD values measured with the
360°. As shown in Figure 7, the uncertainties at
low doses (therefore, low OD) increase. The
larger uncertainties are attributed to concentric
rings that appear in these images.
Improvements in the reconstruction algorithm
are needed to remove these structures from
images. At the 1 Gy dose level, the uncertainty
is lower when using the 360° mode which
suggest that the 360° mode is preferable for
low optical density dosimeters.
Fig. 5 Images of the 4 rod phantom acquired with OCT and xray CT.
0
1.2
0
Comparisons between images of the rod
phantom acquired with OCT and x-ray CT
show very little difference. Overlapping of the
images revealed no discrepancies in the rod
position larger than 1 mm throughout the entire
volume.
Dose (Gy)
The reproducibility of the original
configuration of the OCT system was
determined by imaging the same plane of a
polymer gel dosimeter repeatedly over a 3
hour period. Percent difference images were
computed by comparing the first image with
the other images that were acquired. In
addition, the uncertainty of the OCT system
in measuring optical density values was
determined from the mean value of the
percent difference images. The same
procedure was used to evaluate the fast
scanning platform.
-5
Due to differences in scan speed, the fastscanning platform was able to acquire a
significant larger number of images during a 3
hour period. For both platforms, there was not
a decline in the performance of the scanner
over time.
Fig. 2 X-ray CT image of the phantom used to evaluate the spatial
accuracy of the fast-scanning platform.
Results
-10
Discussion
The uncertainty in measuring optical
densities was 1.0% for the original
configurations and 0.9% for the fastscanning platform.
The fast scanning platform has two modes of
scanning, a 360° mode and a 180° mode.
The 180° mode reduces the scan time in half
by acquiring half the number of projections.
This study evaluated the differences in the
performance of the system when using 180°
mode versus the 360° mode. A batch of
polymer gels irradiated to doses from 1 Gy
up to 6 Gy were scanned using both modes.
Dose response curves were plotted for the
gels. In addition, polymer gels were
irradiated with a half-blocked field to create a
steep dose gradient and images were
acquired using both modes. Profiles were
taken across the steep dose gradient to
evaluate the spatial accuracy of both modes.
-15
Distance From Central Axis (mm)
Fig. 4 Percent difference between the first scan and successive
scans acquired over a 3 hour period using the fast-scanning platform
of the OCT system.
OCT Scans
360 degree mode
1.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Normalized Dose
The spatial accuracy of the system were also
evaluated in this study. A gelatin phantom 17
cm in diameter was manufactured and 4
Teflon® PFA rods 1/8” in diameter were
positioned within the phantom. The phantom
was imaged with x-ray CT and OCT. The two
datasets were registered using external
fiducials and overlapped to identify
differences in the locations of the rods.
Net OD (arbitrary units)
Optical CT systems (OCT) are used to
readout the dose distributions measured by
3D dosimeters. The system measures the
change in optical density of irradiated
dosimeters. The first model of the OCT
system developed by MGS Research Inc.
employed a series of mirrors and a stepping
motor to translate the laser across a
dosimeter to acquire a projection [1]. An
additional stepping motor would turn the
dosimeter after the projection was complete.
This process would repeat until adequate
projections were acquired about 360°.
Recently, MGS Research Inc. made
available a new, fast scanning model that
used a different technique to acquire a
projection. The laser is translated across the
sample using a rotating mirror which
significantly reduces the time to acquire a
projection. Other modifications of the system
include the uses a system of Fresnel lenses
to focus the attenuated laser onto a diffuser.
A photodetector was placed immediately
behind the diffuser to measure the laser
intensity. The new fast scanning platform
was recently installed at the Radiological
Physics Center (RPC). While the advantages
of the upgrades are clear, it is necessary to
demonstrate the performance of the system
is not compromised.
Percent Difference (%)
180 degree mode
18
16
14
12
10
8
6
4
2
0
180 degree mode
360 degree mode
Comparison between the 180° and 360° for
gels irradiated with steep dose gradients also
show very little difference. Agreement between
the profiles shown in Figure 8 is better than 5%
for 97% of the points. The 80%-20% penumbra
width is within 1 mm for both scanning modes.
The fast-scanning platform of the OCT system
shows consistent performance with the
previous model. The system is capable of
significantly reducing scan time making it
possible to image an entire 3D volume in 1.5
hours.
References
1
2
3
4
5
6
Dose (Gy)
Fig. 7 Relative uncertainty in the determination of the OD using 180°
and 360° mode of the fast scanning platform.
1. Gore, J.C., M. Ranade, M.J. Maryanski, and R.J. Schulz,
Radiation dose distributions in three dimensions from
tomographic optical density scanning of polymer gels: I.
Development of an optical scanner. Physics in Medicine &
Biology, 1996. 41(12): p. 2695-704.