807 / 1096
S791
ESTRO 36
_______________________________________________________________________________________________
Material and Methods
Fourteen H&N and ten prostate VMAT plans were created
and their respective QA plans were developed using
Monaco 5.1 treatment planning system and delivered by
an Elekta Synergy Linac equipped with an Agility 160 MLC
system. The cylindrical phantom Octavius 4D was used to
measure the dose distribution.
The 2D-array Octavius 729 consists of 729 vented
ionization chambers arranged in a 27x27 matrix with a
spatial resolution of 10mm.The chamber volume is
0.125cm
3
. The Octavius 1500 array has the same layout
and dimensions but with 1405 ionization chambers with a
chamber volume of
0.06cm
3
.
In order to reconstruct and analyze the measured 3D dose
from each plan, the PTW VeriSoft 6.2 patient plan
verification software was used and a volumetric 3D gamma
index analysis (max dose of calculated volume, suppress
dose below 10% of max dose of calculated volume) for both
3%3mm and 2%2mm criteria was performed to compare
and evaluate the measured and calculated doses for both
arrays.
Results
Table I summarizes the results for both cases. The mean
pass rate of global 3D gamma index for all prostate cases
was superior to 99% with 3%3mm and 95% with 2%2mm
criteria. The Octavius1500 achieves higher results for both
criteria. The mean difference was 2.9% for the gamma
2%2mm
and
0.6%
for
the
3%3mm.
For the H&N cases, the mean passing rate was lower than
prostate cases. Similarly, the Octavius1500 obtain better
results for both criteria. The mean difference was 4.2% for
the gamma 2%2mm and 0.8% for the 3%3mm.
Expectedly, the greatest differences between Octavius
1500 and 729 are shown in the gamma 2%2mm criteria.
Figure 1 shows a comparison between the gamma 2%2mm
for both detectors for a H&N case.
Conclusion
As we expected, the Octavius 1500 achieves a better
result for pre-treatment VMAT plan verification and the
results are more remarkable for the gamma 2%2mm
criteria. In addition, plans with a higher complexity such
H&N can benefit from the superior results of the Octavius
1500.
Moreover, the Octavius 1500 present the possibility to
increase the spatial sampling frequency and the coverage
of a dose distribution by merging two measurements. We
can conclude that Octavius 1500 outperformed Octavius
729 for VMAT pre-treatment QA.
EP-1480 Patient-specific QA for CyberKnife MLC plans
using Monte Carlo
P.H. Mackeprang
1
, D. Vuong
1
, W. Volken
1
, D. Henzen
1
, D.
Schmidhalter
1
, M. Malthaner
1
, S. Mueller
1
, D. Frei
1
, D.M.
Aebersold
2
, M.K. Fix
1
, P. Manser
1
1
Division of Medical Radiation Physics and Department of
Radiation Oncology, Inselspital- Bern University Hospital-
and University of Bern, Bern, Switzerland
2
Department of Radiation Oncology, Inselspital- Bern
University Hospital- and University of Bern, Bern,
Switzerland
Purpose or Objective
To implement and validate a Monte Carlo (MC) based dose
calculation framework to perform patient-specific quality
assurance (QA) for multi-leaf collimator (MLC) based
CyberKnife treatment plans.
Material and Methods
In order to calculate dose distributions independently
from the treatment planning system (TPS), an
independent dose calculation (IDC) framework was
developed based on the EGSnrc MC transport code. The
framework uses XML-format treatment plan input and
DICOM format patient CT data, an MC beam model using
phase space files, CyberKnife MLC beam modifier
transport using the EGS++ class library, a beam sampling
and coordinate transformation engine and dose scoring
using DOSXYZnrc.
Validation of the framework was performed against dose
profiles of single beams with varying field sizes measured
with a diode detector in a water tank in units of cGy /
monitor unit (MU) and against a two-dimensional dose
distribution of a full treatment plan measured with
Gafchromic EBT3 (Ashland Advanced Materials,
Bridgewater NJ) film in a homogeneous solid water slab
phantom. The film measurement was compared to IDC by
gamma analysis using 1% (global) / 1 mm criteria and a 10%
global low dose threshold.
Finally, the dose distribution of a clinical prostate
treatment plan was calculated and compared to dose
calculated by the TPS finite size pencil beam algorithm by
gamma analysis using either 2% (global) / 2 mm or 1%
(global) / 1 mm criteria and a 10% global low dose
threshold.
Results
Dose profiles calculated with the developed framework in
a homogeneous water phantom agree within 3% or 1 mm
to measurements for all field sizes. 87.1% of all voxels pass
gamma analysis comparing film measurement to
calculated dose. Gamma analysis comparing dose
calculated by the framework to TPS calculated dose for
the clinical prostate plan showed 99.9% passing rate for 2%
/ 2 mm criteria and 85.4% passing rate for 1% / 1 mm,
respectively. Dose differences of up to ±10% were
observed in this case near bony structures or metal
fiducial markers.
Conclusion
An MC based modular IDC framework was successfully
implemented and validated against measurements and is
now available to perform patient-specific QA by
independent dose calculation.
EP-1481 Testing algorithms in water and
heterogeneous medium using experimental designs
S. Dufreneix
1
, A. Barateau
1
, M. Bremaud
1
, C. Di Bartolo
1
,
C. Legrand
1
, J. Mesgouez
1
, D. Autret
1
1
Institut de Cancérologie de l'Ouest, Medical Physics,
Angers, France
Purpose or Objective
The IAEA Tecdoc 1580 and 1583 suggest several beam
configurations for testing, commissioning and ongoing
quality assurance of TPS. However, the large number of