Q

u
e
e
n
s P
rin
te
r an
d
C
o
n
tro
lle
r o
f H
M
S
O
, 2
0
1
4
. 1
0
9
9
0
/0
8
1
4
www.npl.co.uk
The purpose of this work is to develop and validate a reliable
method for characterizing small elds on radiotherapy
machines with radiochromic lm. A novel multichannel
analysis method is proposed that focuses on correcting lm
systematic errors introduced by manufacturing imperfections.
To validate the technique, a controlled experimental
environment is used. Initial results show that the performance
of the method is promising, suggesting potential future for
small eld dosimetry.
Methods:
A. Experimental set up
A Cobalt-60 beam
is used to irradiate
radiochromic flms
(Gafchromic
EBT-3) homogenously,
allowing the
determination of
parameters for the flm
scanner homogeneity
correction,
multichannel analysis
and flm dose response
calibration. The setup is shown in Figure 1, while the homogeneity of a
profle across the flm at an SSD of 430 cm in Figure 2.
B. Multichannel method
Prior to performing the multichannel analysis, a correction is applied to
the flm to account for the scanner light inhomogeneity (see Figure 3). A
multichannel method developed from frst principles and based on the
statistical behaviour of optical density for a given dose is then applied
to correct for defects in flm emulsion thickness uniformity. This method
allows the flm response to be characterised in terms of dose and thickness
perturbation, as shown in Figure 4. By applying the multichannel correction,
as shown in Figure 6, a perturbation and dose index map was extracted, thus
allowing the image to be corrected and the flm calibration to dose to be
performed (see Figure 5).
Validation of the method:
The performance of the method is validated against experimental
measurements of dose distributions in a Cobalt-60 beam. Beam profles
(Figures 7 and 9), as well as percent depth dose distributions (Figures 8 and
10), are measured and used to compare the proposed method with one
using single channel analysis (red only). Results with the new method show
a signifcant improvement compared to the single channel method, with
systematic errors being reduced by a factor of up to 3.
Conclusion and future work:
The proposed method based on multichannel analysis is promising for
applications requiring high accuracy, such as in the characterization of small
felds. Future work requires further validation of the method against other
detectors as well as Monte Carlo calculation.
References:
1) Alfonso et al. A new formalism for reference dosimetry of small and nonstandard
felds. Med. Phys. 35 (2008);
2) Bouchard et al. On the characterization and uncertainty analysis of radiochromic
flm dosimetry, 36 (2009);
3) Chung et al. Investigation of three radiation detectors for accurate measurement
of absorbed dose in nonstandard felds, 37 (2010);
4) Micke et al. Multichannel flm dosimetry with nonuniformity correction,
Med. Phys. 38 (2011);
5) Saur and Frengen, GafChromic EBT flm dosimetry with fatbed CCD scanner
Med. Phys. 35 (2008)
Figure 1: Cobalt-60 flm irradiation set up.
TOWARD ACCURATE SMALL PHOTON FIELD
DOSIMETRY USING RADIOCHROMIC FILM
Billas I, Silvestre I, Bouchard H
Radiation Dosimetry Group, National Physical Laboratory, Hampton Road, Teddington, Middlesex, TW11 0LW, UK
ilias.billas@npl.co.uk
Figure 2: Rawsignal profle from a homogenously
irradiated flm.
Figure 3: Scanner homogeneity correction matrix.
Figure 4: 2Dcharacterization of the flm response to dose
and thickness perturbation.
Figure 5: EBT-3 flm response curve.
Figure 6 : Image, thickness perturbation and dose index map.
Figure 7: Cobalt-60 beam profle comparison between single and multi channel analysis (horizontal
on the left and vertical on the right).
Figure 8: Cobalt-60 PDDprofle comparison between single and multi channel analysis.
Figure 9: A 10 x 10 cm
2
Cobalt-60 dose distribution. Figure 10: 2Dimages of a Cobalt-60 PDD.