156 / 1082
S143
ESTRO 36 2017
_______________________________________________________________________________________________
HDR prostate plan. The fluctuation could be reduced to
<0.5% by mixing the Y
2
O
3
:Eu and YVO
4
:Eu phosphors in a
ratio 1-to-10. The stem signal of the ruby, Y
2
O
3
:Eu and
YVO
4
:Eu ISDs was up to 3%, 1% and 2%, respectively, of the
total signal, and the photoluminescence was <1%, when
the BT source moved 8 cm away from the detector and 1
cm from the fiber-optic cable. In contrast, the stem signal
of the PSD was up to 70%.
Conclusion
Red-emitting ISDs based on ruby, Y
2
O
3
:Eu and YVO
4
:Eu are
suitable for HDR BT treatment verification in real time.
Their large signal intensities and emission properties
facilitate accurate detector systems that are
straightforward to manufacture and use which can result
in widespread dissemination and improved patient safety
during BT.
OC-0279 Removing the blindfold - a new take on real-
time brachytherapy dosimetry
J. Johansen
1
, S. Rylander
1
, S. Buus
1
, L. Bentzen
1
, S.B.
Hokland
1
, C.S. Søndergaard
1
, A.K.M. With
2
, G.
Kertzscher
3
, C.E. Andersen
4
, K. Tanderup
1
1
Aarhus University Hospital, Department of oncology,
Aarhus C, Denmark
2
Örebro University Hospital, Department of Medical
Physics, Örebro, Sweden
3
The University of Texas MD Anderson Cancer Center,
Department of Radiation Physics, Houston- TX, USA
4
Technical University of Denmark, Center for Nuclear
Technologies, Roskilde, Denmark
Purpose or Objective
Although in-vivo dosimetry has been available for decades
it is still not a standardized verification tool in
brachytherapy (BT). Major limitations are that in-vivo
dosimeters only provide point dose information and that
the steep dose gradient leads to strong positional
dependency. The aim of this study is to examine whether
it is possible to utilise in-vivo dosimetry for evaluation of
the implant geometry during irradiation in addition to post
hoc dose verification.
Material and Methods
This study includes in-vivo dosimetry measurements from
22 HDR BT procedures for prostate cancer. Needles were
placed in the prostate guided by TRUS with a subsequent
T2W MRI with 2mm slice thickness for treatment planning.
Dose rates were measured using a fiber-coupled Al
2
O
3
:C
luminescent crystal placed in a dedicated needle in the
prostate.
The dose measurements were analysed retrospectively.
The total accumulated dose was compared to the
predicted dose. Secondly, the measured dose rate
originating from each dwell position in a needle was
compared to the predicted dose rate obtained from the
dose planning system. The discrepancies between
measured and predicted dose rates were assumed to be
caused by geometrical offsets of the needles relative to
the dosimeter from the treatment plan. An algorithm
shifted each treatment needle virtually in radial and
longitudinal directions relative to the dosimeter until
optimal agreement between the predicted and measured
dose rates was achieved.
Results
Table 1 shows the relative difference between the
measured and predicted accumulated dose and the
average radial and longitudinal shifts of 337 needles in 22
treatments. The average shifts are expected to
correspond to systematic uncertainties in dosimeter
positions, and the standard deviations reflect the shift of
needles relative to the dosimeter. Two treatments were
not further analysed because of dosimeter drift by
>15mm.
The longitudinal and radial shifts of each needle are
plotted in Fig. 1. The relative needle-dosimeter geometry
was determined with sub-millimetre precision for 98% of
the treatment needles (error bars in Fig. 1). More than 90%
of the needles were shifted less than 4mm longitudinally
and 2mm radially, which is consistent with typical
uncertainties in needle and dosimeter reconstructions and