EDR2 film dosimetry for IMRT verification using low-energy photon ...

EDR2 film dosimetry for IMRT verification using low-energy photon filters
Inhwan Jason Yeo, a) Akbar Beiki-Ardakani, Young-bin Cho, Mostafa Heydarian,
Ting Zhang, and Mohammad Islam
Princess Margaret Hospital, University of Toronto, Toronto, Ontario M5G 2M9, Canada
Received 31 December 2003; revised 21 April 2004; accepted for publication 21 April 2004;
published 11 June 2004
Recently the EDR2 extended dose range film has been introduced commercially for applications
in radiation therapy dosimetry. In addition to characterizing the wide dynamic range, several authors
have reported a reduced energy dependence of this film compared to that of X-Omatic
Verification XV films for megavoltage photon beams. However, those investigations were performed
under limited geometrical conditions. We have investigated the dosimetric performance of
EDR2 film for the verification of IMRT fields at more clinically relevant conditions by comparing
the film doses with the doses measured with an ion chamber and XV films. The effects of using a
low energy scattered photon filter on EDR2 film dosimetry was also studied. In contrast to previous
reports our results show that EDR2 film still exhibits considerable energy dependence a maximum
discrepancy of 9%, compared with an ion chamber at clinically relevant conditions 10 cm depth
for IMRT fields. However, by using the low-energy filters the discrepancy is reduced to within 3%.
Therefore, EDR2 film, in combination with the filters, is found to be a promising two-dimensional
dosimeter for verification of IMRT treatment fields. © 2004 American Association of Physicists in
Medicine. DOI: 10.1118/1.1760190
Key words: EDR2 film, ion chamber, filtration, XV film, dosimetry, IMRT
I. INTRODUCTION
Radiographic film is widely used for verifications of twodimensional
dose distributions in IMRT, stereotactic radiotherapy,
and tomotherapy. 1–11 The most popular radiographic
films for radiotherapy verification includes X-Omatic Verification
XV film and EDR2 extended dose range film, both
manufactured by Kodak, Inc. The former has widely been
used, while EDR2 film has only recently been tested and
used for IMRT. 11–13 The two films are different in their responses
to dose. The difference originates from their differences
in the content of silver–bromide crystals and grain
size: the grain size of EDR2 film is smaller and it contains
less silver bromide. 6 Although the EDR2 film has been designed
for higher dynamic range in dose, it has lower signal
to noise ratio at doses below 10 cGy. 13 In only two reports
comparative studies between the performance of the two film
types and that of a standard ion chamber have been described.
In a plane parallel to the incident beam i.e., depth
dose measurement, Olch 13 showed good agreement of
EDR2 film results i.e., 2% over-response with those of an
ion chamber, while the results of XV film was shown to
differ by as much as 15% compared to the dose measured by
an ion chamber. Esthappan et al. 12 investigated the difference
in the sensitivities of these two films under different
scattering conditions by measuring the changes of the characteristic
curves as a function of depth. The study reported
little variation in the characteristic curves of the two films
with depth changes. The apparent discrepancy of the results
of the earlier two studies and for XV film is likely due to the
differences in measurement conditions: Esthappan et al. 12
has considered the depth changes down to only 15 cm for the
maximum field size of 15 cm15 cm, while Olch 13 has considered
the depth down to 25 cm for the maximum field size
of 20 cm20 cm. The proportion of low energy scattered
photons has increased as depth and field size increase.
In a plane perpendicular to the beam axis, on the other
hand, both investigators reported good agreement between
film and ion chamber measurement. Esthappan et al. 12 described
the dosimetric study for a 15 cm15 cm field in an
outside-penumbra region and reported good agreement for
EDR film with measurement at 10 cm depth, while Olch 13
reported agreement within 1% for both XV and EDR2 films
at a location 1.5 cm outside the field edge of a 10 cm
10 cm field and at a depth of 1.5 cm.
In this study we have investigated the dosimetric performance
of EDR2 film for IMRT field verification. For comparative
study, dose measurements have been performed at a
depth of 10 cm for specially designed IMRT fields 5 with
EDR2 film, XV film, and an ion chamber. In addition, the
effect of a low-energy (400 keV) photon filter 5,14,15 on the
performance of EDR2 film on IMRT field dosimetry was
studied. The fields were designed to simulate intensity modulation,
which may provide more clinically relevant beams
involving a greater amount of scattering than that of the experimental
conditions used by previous investigators. 12,13
II. METHOD
In this study, two different setups were employed: a perpendicular
setup where film is placed orthogonally to the
beam axis and a parallel setup where film is in parallel with
the beam axis. 5,14,15 In the parallel setup, the film and phantom
were laterally compressed by external force in a compression
device, while in the perpendicular setup they were
vertically compressed by the weight of the phantom. Film
1960 Med. Phys. 31 „7…, July 2004 0094-2405Õ2004Õ31„7…Õ1960Õ4Õ$22.00 © 2004 Am. Assoc. Phys. Med. 1960

1961 Yeo et al.: EDR2 film dosimetry with filters for IMRT 1961
TABLE I. Calibration setup used in this study. * an additional film for background is needed.
Film and beam
alignment
Filter
MUs for XV*
MUs for EDR2*
SSD
cm
Field size
cm
Measurement
condition
Perpendicular
No
5,10,15,20,25,30,35,40,45,50
5,10,20,50,100,150,200,250
100 66 10 cm depth
calibration curves were determined under calibration conditions
which are summarized in Table I. No additional calibration
was performed with the filters in place. This is because
the amount of scattering and thus the effect of filtration
are minimal (2%, a relatively small value compared with
the film error at beam axis of the small field size used in this
study. The film setup is similar to those employed in our past
studies and 0.15-mm-thick lead filters were placed at 6 mm
distance from the film on both sides. 5,14,15
IMRT beams which are measurable by an ion chamber
have been designed to test the dosimetric response of the two
films: the two types of IMRT beams produce pyramid and
inverse-pyramid density shapes on the film. They are similar
to those we have previously employed but the current beams
involve more scattering: A graphical representation of these
beams was previously given. 5 Segment combinations of
these beams are shown in Tables II and III, respectively. The
fields simulates intensity modulation, which provides a
greater amount of scattering than the experimental conditions
used by previous investigators. 12,13 The monitor units and the
minimum field size were carefully chosen to avoid film saturation
and to ensure electronic equilibrium. Parallel exposures
with and without the filters in place were performed as
they have been used consistently in the past for the evaluation
of film response. 5,14,15
The transverse dose profile of each segment in the tables
was measured at 10 cm depth in a water phantom using a
0.125 cm 3 ion chamber IC10. Charge collection for the
same number of MUs was performed on the central axis at
10 cm depth for each segment of the pyramid beam in Table
II. For the segments of the inverse-pyramid beam shown in
Table III, the charge was collected at 8.3 cm off-axis distance
or X1 in the dose plateau region at the depth of 10 cm. The
dose to MU ratio i.e., cGy/MU was first obtained for segment
E of the pyramid beam using the output factor and the
percentage depth dose for the given conditions i.e., field size
and SSD. Dose to MU ratios for the other segments were
individually obtained by multiplying the earlier ratio for segment
E by the ratio between the collected charges of each
segment and segment E. For each segment, these ratios were
then multiplied by the delivered MUs to generate the absolute
dose value at the measurement point. This value was
used for the normalization of the measured relative dose profile
of the corresponding segment in order to determine the
absolute dose profile. By adding together the absolute dose
profiles of all involved segments in the pyramid and inversepyramid
beams, their corresponding composite dose profiles
were created.
Dose calculations were also performed using a treatment
planning system TPS Cadplan, Varian, Inc. with the same
beam configurations and compared with the measured results.
This was done because the film dosimetry is normally
compared with the TPS calculation and thus the evaluation
of the accuracy of the TPS relative to ion-chamber measurement
is necessary.
III. RESULT AND DISCUSSIONS
Figure 1 shows the polynomials of the characteristic
curves of the two films that fit the measured data obtained
using the condition described in Table I. The curves were
used to convert the optical densities of the exposed films to
the dose.
Figures 2 and 3 show the profiles of the pyramid and
inverse-pyramid beams, respectively, measured by EDR2
film. Figures 4 and 5 show the corresponding results for XV
film. In the high dose region of the pyramid field, EDR2 film
showed an over-response of a few percent from the standard
ion chamber measurement, while XV film showed good
agreement. Moreover, there is a negligible effect of the filtration
for the two films in the high dose region. This can be
explained by the fact that the high dose region is on the
central axis under these exposure conditions Table II,
where a minimal amount of scattered photons are present.
On the other hand, in the high dose region of the inversepyramid
field, the two films over-responded by approximately
5%–6%, relative to the dose maximum. The effect of
filtration in this case was significant: it resulted in 4% underresponse
for XV film and 2% over-response for EDR2 film.
The fact that the film error is greater for the inverse-pyramid
beam than for the pyramid beam and the filtration effect was
significant for the inverse-pyramid field can be simply explained
by the difference between the two conditions: the
difference is characterized by the fact that the high dose re-
TABLE II. Irradiation conditions for the pyramid field. Y
(or in-plane field length)21 cm. Gantry and collimator at 0°. Energy
6 MV. SSD100 cm. 10 MUs was used for each field.
Field size A B C D E
X1/X2 3/3 4.5/4.5 6/6 7.5/7.5 9/9
TABLE III. Irradiation conditions for the inverse-pyramid field. Y
(or in-plane field length)21 cm. Gantry and collimator at 0°. Energy
6 MV. SSD100 cm. 7 MUs were used for each field.
Field
size A B C D E F G
X1/X210.5/4.5 10.5/3.0 10.5/1.5 10.5/10.51.5/10.53/10.54.5/10.5
Medical Physics, Vol. 31, No. 7, July 2004

1962 Yeo et al.: EDR2 film dosimetry with filters for IMRT 1962
FIG. 1. Characteristic curves for XV and EDR 2 films.
FIG. 4. The pyramid beam profiles measured by XV film at the depth of 10
cm.
FIG. 2. The pyramid beam profiles measured by EDR2 film at the depth of
10 cm.
gion of the inverse-pyramid field is affected by the outside
penumbral irradiation of segments, while that of the pyramid
field is not.
In the outside penumbra region of the pyramid and
inverse-pyramid fields and the central part local minimum
of the inverse-pyramid field where the low-energy photons
are abundant, the over-response of the two films is about 9%
of the maximum dose. This corresponds to a significant overresponse
of as much as 100% relative to the local dose in the
above areas. Significant reduction down to 3% was
achieved by using the filers of the low-energy photons. This
reduction which was previously shown for relative dosimetry
using XV film 5 has been reproduced in this study for absolute
dosimetry using the two films. However, the observed
over-response of EDR2 film is different from or may
complement the finding of the previous studies 12,13 discussed
in the introduction. The reason is in the differences
between the measurement condition employed in this study
and that in the previous studies. Measurement of the dose
profiles in this study was at 10 cm depth, which is much
FIG. 3. The inverse-pyramid beam profiles measured by EDR2 film at the
depth of 10 cm.
FIG. 5. The inverse-pyramid beam profiles measured by XV film at the
depth of 10 cm.
Medical Physics, Vol. 31, No. 7, July 2004

1963 Yeo et al.: EDR2 film dosimetry with filters for IMRT 1963
greater and clinically more relevant than that used by Olch. 13
In addition, the field size we used is greater than that employed
by Esthappan. 12 Additionally with several steps of
intensity modulation, the radiation conditions in this study
are associated with more scattered photons (400 keV). As
a result, EDR2 film tends to respond less accurately within
the IMRT field area due to the presence of high density silver
bromide.
Whereas EDR2 film was reported to be an accurate dosimeter
for depth dose measurement by the past two
studies, 12,13 it is shown to be not as accurate for IMRT profile
measurement by this study. This makes it interesting to note
the different finding on EDR2 film between the two planes of
measurement parallel and perpendicular to the beam. The
difference between the two planes is attributed to the difference
in the radiation conditions of the two planes characterized
by the proportion of the low-energy photons. In the
parallel plane, the amount of the low-energy photons increases
as depth increases, but their proportion in an entire
spectrum is still influenced by the higher energy part of the
spectrum. On the contrary, in the perpendicular plane at
some sufficient depth beyond the depth of the dose maximum
the low-energy photons are mainly present outside the
penumbra region, and their effect are dominant there. Therefore,
the contribution of the low-energy photons to the film
response is more pronounced in the profile measurement
outside penumbra at deep parts of a phantom. This appears
to be the most reasonable explanation that hardly makes the
change made in the silver–bromide content of EDR2 film
effective in enhancing the dosimetric properties of the film
i.e., less sensitive to the low-energy photons for profile
measurement.
The TPS calculations agree well with ion chamber measurement
in the high dose region. However, they show an
under-response of around 3%, compared with the ion chamber
measurement, in the outside-penumbra regions and the
local minimum of the inverse-pyramid field, while it shows
agreement in the outside penumbra regions of the pyramid
field. This discrepancy is due to the fact that the points in the
areas of the outside-penumbra and the local minimum of the
inverse-pyramid field is closer to the edges of the beamlets
than that of the pyramid field is. Therefore, the known underresponse
of the dose calculation of the TPS in outside penumbra
regions 16 has contributed to the response to the
inverse-pyramid field.
IV. CONCLUSION
While for the measurement in a parallel plane to the beam
axis EDR2 film is reported to be more accurate than XV film,
the former is comparable to the latter for the measurement in
a perpendicular plane for IMRT beams which involve great
amount of low-energy scattered photons. Similarly to XV
film, the use of filters significantly reduces errors in EDR2
film dosimetry and results in an agreement with the ion
chamber to within 3%. This is a promising result considering
the limited accuracy of treatment planning calculations
to which film measurement is compared clinically and the
current difficulty to measure the dose distribution from an
IMRT beam with both accuracy and fine resolution. The use
of the filters for the profile measurement, together with the
proven accuracy for depth dose measurement, 13 enables accurate
dosimetry using EDR2 film in nearly any measurement
point throughout a phantom. Film dosimetry can be
quantitatively performed based on the evaluated errors of
both film dosimetry and TPS calculation.
a Author to whom correspondence should be addressed; electronic mail:
medicphys@hotmail.com
1 B. Paliwal, W. Tomé, S. Richardson, and T. R. Mackie, ‘‘A spiral phantom
for IMRT and tomotherapy treatment delivery verification,’’ Med.
Phys. 27, 2503 2000.
2 M. Oldham and S. Webb, ‘‘Intensity-modulated radiotherapy by means of
static tomotherapy: A planning and verification study,’’ Med. Phys. 24,
827–836 1997.
3 J. N. Yang, T. R. Mackie, P. Reckwerdt, J. O. Deasy, and B. R. Thomadsen,
‘‘An investigation of tomotherapy beam delivery,’’ Med. Phys. 24,
425–436 1997.
4 T. D. Solberg, F. E. Holly, De Salles, A. F. Antonio, R. E. Wallace, and J.
B. Smathers, ‘‘Implications of tissue heterogeneity for radiosurgery in
head and neck tumors,’’ Int. J. Radiat. Oncol., Biol., Phys. 32, 235–239
1995.
5 S. G. Ju, Y. C. Ahn, S. J. Huh, and I. J. Yeo, ‘‘Film dosimetry for intensity
modulated radiation therapy,’’ Med. Phys. 29, 351–355 2002.
6 I. J. Yeo, C. Wang, and S. E. Burch, ‘‘A new approach to film dosimetry
for high-energy photon beam using organic plastic scintillators,’’ Phys.
Med. Biol. 44, 3055–3069 1999.
7 J. F. Williamson, ‘‘Film dosimetry of megavoltage photon beams: A practical
method of isodensity-to-isodose curve conversion,’’ Med. Phys. 8,
84–98 1981.
8 J. Purdy, ‘‘Instrumentation for dosimetric measurements,’’ Proceedings of
the AAPM Conference on Electron Linear Accelerators in Radiation
Therapy, 1978.
9 D. A. Low, C. Chao, S. Mutic, R. L. Gerber, C. A. Perez, and J. A. Purdy,
‘‘Quality assurance of serial tomotherapy for head and neck patient treatments,’’
Int. J. Radiat. Oncol., Biol., Phys. 42, 681–692 1998.
10 D. A. Low, S. Mutic, J. F. Dempsey, R. L. Gerber, W. R. Bosch, C. A.
Perez, and J. A. Purdy, ‘‘Quantitative dosimetric verification of an IMRT
planning and delivery system,’’ Radiother. Oncol. 49, 305–316 1998.
11 X. R. Zhu, P. A. Jursinic, D. F. Grimm, F. Lopez, J. J. Rownd, and M. T.
Gillin, ‘‘Evaluation of Kodak EDR2 film for dose verification of intensity
modulated radiation therapy delivered by a static multileaf collimator,’’
Med. Phys. 22, 1687–1692 2002.
12 J. Esthappan, S. Mutic, W. B. Harms, J. F. Dempsey, and D. A. Low,
‘‘Dosimetry of therapeutic photon beams using an extended dose range
film,’’ Med. Phys. 29, 2438–2445 2002.
13 A. J. Olch, ‘‘Dosimetric performance of an enhanced dose range radiographic
film for intensity-modulate radiation therapy quality assurance,’’
Med. Phys. 29, 2159–2168 2002.
14 S. E. Burch, K. J. Kearfott, J. H. Trueblood, W. C. Shiels, I. J. Yeo, and
C. K. Wang, ‘‘A new approach to film dosimetry for high energy photon
beam: Lateral scatter filtering,’’ Med. Phys. 24, 775–783 1997.
15 I. J. Yeo, S. E. Burch, and C. K. Wang, ‘‘A filtration method for improving
x-ray film dosimetry in photon radiotherapy,’’ Med. Phys. 24, 1943–
1953 1997.
16 P. R. M. Storchi, L. J. van Battum, and E. Woudstra, ‘‘Calculation of a
pencil beam kernel from measured photon beam data,’’ Phys. Med. Biol.
44, 2917–2928 1999.
Medical Physics, Vol. 31, No. 7, July 2004