ESTRO 36 Abstract Book

911 / 1096

S895 ESTRO 36
_______________________________________________________________________________________________
Material and Methods
Dataset for a total of 945 CBCT scans were obtained from
30 patients (13 with H mask and 17 with HS mask). For
each fraction, patients were placed on the HexaPOD
robotic couch and a CBCT was performed. For
interfraction accuracy evaluation, the daily variations of
the three principal axes (X, Y and Z) and three rotational
movements (pitch, roll, and yaw) were extracted. Also,
the type of thermoplastic mask was recorded. The
following parameters were calculated: the mean of the
setup corrections (M), the standard deviation (random
error, σi), the group random error (σ) defined as the mean
of all the individual patient random error σi and the
systematic group error (∑) defined as the standard
deviation of all the means measured for each patient.
Results
Results are shown in Table I. The overall mean
displacements are larger for the HS mask patients than for
the H mask. Also a trend toward a positive antero-
posterior direction (1.19 mm) was observed in
translational displacements for the HS mask patients. This
could be explained taking into account the differences
between the simulation and treatment couches as well as
the accuracy of the room lasers. The remaining error
components did not show any trend.
For the rotational directions, the bigger error was in the
pitch direction for both the H mask (0.76) and HS mask
patients (0.85).
Regarding the systematic group error, we have found a
larger error for the HS mask patients (2.34) again toward
a positive antero-posterior direction. The random group
error shows the same behavior as well.
Conclusion
We can conclude that the patients with HS mask have
larger displacements than the H mask patients, but further
investigations should be done.
All errors are below 3mm. This result agrees with the
literature for H&N displacements.
In summary, HexaPOD couch in combination with daily
CBCT can considerably improve the accuracy of patient
positioning during VMAT treatment for H&N treatments.
EP-1649 Comparison of two thermoplastic
immobilization shells for frameless stereotactic
radiotherapy
I. Gagne1 , S. Zavgorodni1, A. Alexander2, I. Vallieres2
1 BC Cancer Agency - Vancouver Island, Medical Physics,
Victoria, Canada
2 BC Cancer Agency - Vancouver Island, Radiation
Oncology, Victoria, Canada
Purpose or Objective
The primary goal of this study was to compare the inter-
fraction 6DOF corrections applied between patients
immobilized with an open-face shell system and patients
immobilized with a reinforced closed-shell system. The
intra-fraction motion was also compared between these
two groups of patients.
Material and Methods
Sixty patients have been treated with frameless
radiotherapy on a Varian TrueBeam STx linear accelerator
equipped with a six-degree of freedom (6DOF) couch. All
patients had a planning CT scan with an immobilization
system that comprised of a CIVCO head cup, a CIVCO
customizable pillow and a thermoplastic shell. An open-
faced shell from CIVCO was used on the first 15 patients
in anticipation of the ALIGN RT optical tracking system
installation. A closed-faced shell from AUBO with extra
reinforcement was employed on subsequent patients due
to delays in the approval of ALIGN RT in Canada and larger
than expected setup and intra-fraction motions with the
open-faced immobilization system. Two pre-treatment
CBCTs were acquired; the first to correct using 6DOF bone
anatomy matching the initial inter-fraction setup error,
the second to correct using 4DOF the residual setup error
following the 6DOF couch moves. A post CBCT was
acquired to determine the intra-fraction motion using
6DOF bone anatomy matching.
Results
Datasets from 12 patients with an open-faced
immobilization shell and 29 patients with a reinforced
closed-face immobilization shell were obtained for a total
of 2451 CBCT scans. Table 1 summarizes the population
average of the patient average and largest 6DOF
corrections (vrt, lng, lat, pitch, roll, rtn) from CBCT1 for
both the open-face and closed-face shells. Also included
is the population average of the average and largest intra-
fraction motion (vrt, lng,lat,pitch, roll, rtn) recorded from
CBCT3 for both the open-face and closed-face shells. A
student t-test for uneven sample and variance was applied
to determine which parameters had statistically
significant differences at the p=0.01 level. No statistically
significant differences were found between the two
patient populations when patient 6DOF correction
averages, however, statistically significant differences
were found between the two patient populations when the
patient largest rotational corrections were used in the
test. Statistically significant differences between the two
patient populations were also noted for the patient largest
6DOF pitch and roll intra-fraction motion as well as the
patient average 6DOF pitch intra-fraction motion.
Conclusion
Moving to a reinforced closed-face immobilization shell
from an open-faced immobilization system has
significantly reduced the magnitude of the rotational
corrections as well as significantly reduced the magnitude
of the pitch and roll motion errors.
EP-1650 Setup uncertainty in head and neck assessed
by a 1.5T MR-sim with thermoplastic mask
immobilization
Y.H. Zhou1, W.W.K. Fung2, J. Yuan 1, O.L. Wong1, G.
Chiu2, K.F. Cheng 2, K.Y. Cheung1, S.K. Yu 1
1Hong Kong Sanatorium & Hospital, Medical Physics &
Research Department, Happy Valley, Hong Kong SAR
China
2Hong Kong Sanatorium & Hospital, Department of
Radiotherapy, Happy Valley, Hong Kong SAR China