970 / 1082
S954
ESTRO 36 2017
_______________________________________________________________________________________________
device (EPID) and make this process as fast and accurate
as possible.
Material and Methods
The LINAC is an Elekta Synergy with Agility MLC and 6 MV
photons. A software is developed in MATLAB with some
remarkable points:
1.
Elekta iCOMCAT software was employed to
generate and send the strip-test with multiple
segments as a unique treatment, as is much
faster than creating and irradiating a beam for
each segment. With the software of Elekta
iView is difficult to acquire a complete image
of each full segment as this is not fast enough,
so fluency corrections of these segments were
performed, in order to avoid erroneous pixel
values (PV) in the way: a) In a 23x23 open field
is acquired a horizontal profile and measure
the % PV (in the center position of each future
segment), this % is related to the PV of the
position of a reference segment. b) Measure
the mean PV in the center of each strip-test
segment, and obtain the % PV related to the
reference segment PV. c) Rescale the image of
each segment in order to obtain the % PV
(respect the reference segment). Finally make
the sum of all images.
2.
Segments of 2 x 20 cm (cross-plane x in-plane)
to form series of strip-test images with gaps
overlapping from 1.2 to 3 mm are acquired for
taking the MLC reference after calibration. The
strip-test need bigger gap spread than other
MLC in order to detect the gap position
correctly, because of the lower penumbra.
3.
To correct the collimator angle is used the
filtered back projection method, because is very
tricky to use the interleaf leakage, as this MLC
have much lower interleaf transmission than
other MLC, like Millenium (from Varian).
4.
To localize the radiation center (RC) of the EPID
is used a LINAC tray with centered radiopaque
mark. Four 20x20 fields are obtained with this
tray at 4 collimator angles. RC is determined for
gantry 0º detecting the mark position in each
image and obtaining the mean. A vector
displacement is created to obtain RC with one
image at 0º collimator. Tray images for various
gantry angles at 0º collimator are acquired, so
that with just one tray image is enough to detect
RC exactly. This method is faster than using
field edges, where at least 2 images at different
collimator angles must be acquired for each
gantry angle.
Measurements of leaf positions using light projection are
made. Also are obtained strip-test with films and analyzed
with
RIT
software.
Results
The differences in leaf positions compared with film and
light field are beyond 0.1 mm and 1 mm (light field edge
detection has a much bigger uncertainty). The acquisition
and analysis for one strip-test take less than 4 min.
Conclusion
The methodology employed analyzes a MLC strip-test in an
Elekta LINAC in a fast and accurate way.
EP-1758 Towards Clinical Implementation of an Online
Beam Monitoring System
M. Islam
1
, M. Farrokhkish
2
, Y. Wang
2
, B. Norrlinger
2
, R.
Heaton
1
, D. Jaffray
1
1
Princess Margaret Cancer Centre and University of
Toronto, Medical Physics, Toronto, Canada
2
Princess Margaret Cancer Centre, Medical Physics,
Toronto, Canada
Purpose or Objective
Continual advancement of Radiation Therapy techniques
and consequent complexity in planning and delivery
require constant vigilance. To address this, the idea of
independent real-time beam monitoring has been
proposed. In this presentation, we describe initial steps
towards introducing the Integral Quality Monitoring (IQM)
system into clinical practice.
Material and Methods
The IQM system (manufactured by iRT, Germany) consists
of a large-area ion chamber mounted at Linear
Accelerator’s (Linac) accessory slot, which provides a
spatially dependent “dose -area- product” per field
segment. The system monitors beam delivery in real-time
by comparing the expected and measured signals.
Initial evaluation of the system included: reproducibility
and stability, agreements between calculated vs.
measured signals, sensitivity and specificity for errors. A
multiphase approach was considered for clinical
implementation. First, IQM data are collected during
conventional dosimetric QA tests. Results of the QA tests
with and without the IQM chamber in the beam are
compared. During this phase a reference dataset of
measured IQM signals vs. calculated is generated to help
determine the tolerance in the predicated
signals. Second, IQM system is introduced as the primary
pre-treatment QA tool. Gain in work-flow efficiency and