| dc.description.abstract |
<p>Purpose: To develop a comprehensive method to measure the radiation uncertainty
and coincidence with the kV-CBCT imaging coordinate system using NIPAM kV-CBCT dosimetry.</p><p>Methods:
An N-isopropylacrylamide (NIPAM) dosimeter is irradiated at eight gantry/couch combinations
which enter the dosimeter at unique orientations such that the beams do not overlap
except at the isocenter. 1-3 CBCT images are acquired before and immediately after
irradiation, radiation profile is detected per beam, and the displacement from the
imaging isocenter is quantified. This test has been performed on SRS cone sizes ranging
from 4 mm to 15 mm diameter and a 5 mm diameter MLC field, delivering approximately
16 Gy per beam. Matlab code was developed in house to detect each beam’s geometry
and to quantify relevant parameters, including radiation isocenter and coincidence
with the CBCT origin and the actual gantry and couch angles per beam. The dose profile
of each beam was detected in the CBCT using the contrast-to-noise ratio (CNR) of the
irradiated high dose regions relative to the surrounding background signal of the
dosimeter. Reproducibility was demonstrated by repeating the test on two separate
NIPAM dosimeters using the 6 mm cone. To determine the robustness of our test, our
results were compared to the results of the traditional Winston-Lutz test, film based
“star shots,” and the Varian Machine Performance Check (MPC). The ability of our Matlab
code to detect alignment errors was demonstrated by applying a 0.5 mm shift to the
MLCs in the direction of leaf travel. </p><p>Results: Setup, irradiation, and imaging
can be completed in under 40 minutes. The minimum radius to encompass all beams calculated
by automated analysis for the MLCs, 4 mm cone, 6 mm cone, 7.5 mm cone, 12.5 mm cone,
and 15 mm cone was 0.38 mm, 0.44 mm, 0.53 mm, 0.48 mm, 0.75 mm, 0.5 mm, and 0.57 mm,
respectively. When determined manually, these values slightly decreased to 0.28 mm,
0.40 mm, 0.33 mm, 0.41 mm, 0.61 mm, 0.48, and 0.34 mm, respectively. The isocenter
verification test was repeated using the 6 mm cone; in both tests, the smallest radius
to encompass all beams was found to be 0.53 mm, indicating that the test is reproducible.
For comparison, the 3D isocenter radius was 0.24 mm, 0.25 mm, and 0.28 mm for the
traditional Winston-Lutz test with MLCs, the Varian MPC, and a “star shot” QA sample.
Lastly, when a 0.5 mm shift was applied to the MLCs, the smallest radius to encompass
all beams increased from 0.38 mm to 0.90 mm.</p><p>Conclusion: The results of this
project demonstrate the feasibility of a comprehensive isocenter verification test
using NIPAM kV-CBCT dosimetry which incorporates the evaluation of radiation coincidence
with the imaging coordinate system, and is capable of producing sub-mm results. This
test is applicable to all SRS cone sizes as well as MLCs and can be performed in a
typical QA time slot.</p>
|
|
