Physics in Medicine & Biology

ACCEPTED MANUSCRIPT

Small field dosimetry correction factors for circular and MLC shaped
fields with the CyberKnife M6 System: evaluation of the PTW 60023
microSilicon detector
To cite this article before publication: Paolo Francescon et al 2019 Phys. Med. Biol. in press https://doi.org/10.1088/1361-6560/ab6154

Manuscript version: Accepted Manuscript

Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process,
and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted
Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors”

This Accepted Manuscript is © 2019 Institute of Physics and Engineering in Medicine.

During the embargo period (the 12 month period from the publication of the Version of Record of this article), the Accepted Manuscript is fully
protected by copyright and cannot be reused or reposted elsewhere.
As the Version of Record of this article is going to be / has been published on a subscription basis, this Accepted Manuscript is available for reuse
under a CC BY-NC-ND 3.0 licence after the 12 month embargo period.

After the embargo period, everyone is permitted to use copy and redistribute this article for non-commercial purposes only, provided that they
adhere to all the terms of the licence https://creativecommons.org/licences/by-nc-nd/3.0

Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content
within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this
article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions will likely be
required. All third party content is fully copyright protected, unless specifically stated otherwise in the figure caption in the Version of Record.

View the article online for updates and enhancements.

Page 1 of 10 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3
4 Small field dosimetry correction factors for circular and MLC
5
6 shaped fields with the CyberKnife M6 System: evaluation of
7
8
the PTW 60023 microSilicon detector

pt
9
10
11
12 P. Francescon1, W. Kilby2*, J. M. Noll2, N. Satariano1, and C. Orlandi1
13

cri
14
15
16
17
18 Author’s addresses: 1Department of Radiation Oncology, Ospedale Di Vicenza, I-36100 Vicenza, Italy,
19 2
Accuray Incorporated, 1310 Chesapeake Terrace, Sunnyvale, CA 94089, USA
20

us
21
*Author to whom correspondence should be addressed: Warren Kilby, Accuray Incorporated, 1310
22
23 Chesapeake Terrace, Sunnyvale, CA 94089, USA. Telephone +44 7881 301885. Email:
24 wkilby@accuray.com
25
26
27
28
29
an
Short title: CyberKnife small field dosimetry corrections for the PTW microSilicon

30
31
32
M
33
34
35
36
37
38
39
d

40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 Abstract
4
5 The PTW 60023 microSilicon is a new unshielded diode detector for small-field photon dosimetry. It
6
7
provides improved water equivalence and a slightly larger sensitive region diameter in comparison to
8 previous diode detectors in this range. In this study we evaluated the correction factors relevant to

pt
9 commissioning a CyberKnife System with this detector by Monte Carlo simulation and verified this
10 data by multi-detector measurement comparison. The correction factors required for output factor
11 determination were substantially closer to unity at small field sizes than for previous diode versions
12 𝑓 ,𝑓
13 (e.g. 𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟
= 0.981 at 5 mm field size which compares with corrections of 5% - 6% with other

cri
14 stereotactic diodes). Because of these differences we recommend that corrections to small field output
15
factor measurements generated specifically for the microSilicon detector rather than generic data taken
16
17 from other diode types should be used with this new detector. For depth-dose measurements the
18 microSilicon is consistent with a microDiamond detector to <1% (global), except at depths <10 mm
19 where the diode gives a significantly lower measurement, by 6% - 8% at the surface. For profile
20 measurements, the microSilicon requires negligible corrections except in the low dose region outside

us
21
22
the beam, where it underestimates off-axis-ratio for small fields and overestimate for large fields. Where
23 this effect is most noticeable at the largest field size and depth (115 mm x 100 mm and 300 mm depth)
24 the microSilicon overestimates OAR by 2.3% (global) in the profile tail. This is consistent with other
25 unshielded diodes.
26
27
28
29
an
30
31
32
M
33
34
35
36
37
38
39
d

40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
Page 3 of 10 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 1. Introduction
4
5 Accurate beam data measurement, including Output Factor (OF), Percent Depth-Dose (PDD), and Off-
6
7
Axis Ratio (OAR), is challenging for the small radiation fields employed in radiosurgery and SBRT
8 (Asprakis et al 2010). A recent formalism proposed the use of detector and treatment beam specific

pt
9 correction factors to convert measurement ratios into the corresponding dose ratios to define these beam
10 data (Alfonso et al 2008, Alfonso et al.2017). We have previously evaluated these corrections for the
11 CyberKnife® Robotic Radiosurgery System (Accuray Incorporated, Sunnyvale, CA) as appropriate for
12
13 stereotactic diodes, small-volume ionization chambers, synthetic microDiamond detectors, and point

cri
14 scintillation detectors (Francescon et al 2012, 2014a, b, 2017). The most commonly used detector for
15 CyberKnife commissioning is currently the PTW 60018 stereotactic diode (PTW-Frieburg, Germany),
16 which the vendor has recently replaced with the 60023 microSilicon detector. Both are p-type silicon
17
diodes designed for use in small photon fields and share the same external form factor. However, there
18
19 are differences in their construction (table 1) that might affect the small field response. For unshielded
20 diodes, field size dependent changes in the perturbation factor associated with detector encapsulation

us
21 around the silicon, which is largely epoxy resin, contributes about 50% of the overall correction factor
22 for small field OF measurements (Francescon et al.2014b), and therefore the lower density of the epoxy
23
used in the microSilicon design might result in a correction factor closer to unity. In addition, the
24
25 increased sensitive region diameter with the microSilicon design will increase the geometric
26 perturbation factor at small field sizes (it increases by about 1% for a CyberKnife 5mm fixed field based
27
28
29
an
on a volume averaging calculation). The purpose of this study was to investigate these differences, and
to provide overall PTW 60023 correction factors for use with CyberKnife.
30
31
32
Table 1. Key differences between the PTW 60018 stereotactic diode and its replacement the 60023
M
33
34 microSilicon diode (nominal values provided by vendor, except 60018 epoxy density from Francescon
35 et al. (2012)). The nominal 60023 sensitive region diameter is in good agreement with experimental
36 determination (Poppinga et al. 2019)
37
38 Diode 60018 Diode 60023
39 Entrance window 0.3 mm RW3 (1.045 g/cm3) 0.3 mm RW3 (1.045 g/cm3)
d

40 0.27 mm epoxy (1.4 g/cm3) 0.01mm Al (2.71 g/cm3)

41 0.48 mm epoxy (1.15 g/cm3)
42
Sensitive region diameter 1.1 mm 1.5 mm
43
pte

Sensitive region thickness 250 µm 18 µm

44
45
Sensitive region volume 0.24 mm3 0.03 mm3
46
47
48 2. Method
49
50 We have previously reported Monte Carlo simulations to generate measurement correction factors, kΩ,
51 for the CyberKnife M6TM System with circular and MLC shaped fields, applied to multiple detector
ce

52 designs including the PTW 60018 (Francescon et al., 2017),

53
𝐷(𝑟,𝑧,𝑓)𝑤
54 ⁄𝐷(𝑟,𝑥,𝑓)
55 𝑘Ω (𝑟, 𝑧, 𝑓, 𝑟𝑟𝑒𝑓 , 𝑧𝑟𝑒𝑓 , 𝑓𝑟𝑒𝑓 ) = 𝐷(𝑟𝑟𝑒𝑓 ,𝑧𝑟𝑒𝑓 ,𝑓𝑟𝑒𝑓 )𝑤
𝑑𝑒𝑡
(1)
⁄𝐷(𝑟 ,𝑧
56 𝑟𝑒𝑓 𝑟𝑒𝑓, 𝑓𝑟𝑒𝑓 )𝑑𝑒𝑡
Ac

57
58 where D(r,z,f) is the MC calculated absorbed dose per history at off-axis position r, depth z, and field
59 aperture f. The subscript “det” indicates that dose is scored within the sensitive volume of the detector
60 placed within a water phantom, and “w” indicates that dose is scored at the same position in the absence
 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 of the detector. With appropriate selection of geometric parameters this method was used to calculate
4 correction factors for OF, PDD, and OAR. Further details are provided in Francescon et al. (2017).
5
6 In this work a model of the microSilicon detector was constructed using details provided by the vendor.
7 This model was used to calculate kΩ for OF over the full range of fixed collimator and MLC field sizes
8 in the manufacturers reference geometry of 785 mm SSD and 15 mm depth, relative to the 60 mm fixed

pt
9 collimator (which is the machine specific reference field). The correction factor kΩ was also calculated
10
for OAR measurements at 785 mm Source-Surface Distance (SSD) for the 5 mm and 60 mm fixed
11
collimators. In all simulations the detector was oriented with its stem parallel to the beam axis as per
12
13
vendor recommendations, in a water phantom at 785 mm SSD.

cri
14
A chain of multiple MC simulations was employed to evaluate the detector perturbations associated
15
with OF correction. This method was first described for ionization chambers by Bouchard et al. (2009).
16
17
This was amended for use with diode detectors by Francescon et al. (2014b), and the perturbation factor
18 nomenclature from that paper are used here.
19
OF measurements were performed using a CyberKnife M6 system at the vendor’s test facility. OF
20
measurements were made with fixed collimators and an InCiseTM 2 MLC (Asmerom et al. 2016) using

us
21
22 PTW 60023 and 60018 diodes, PTW 60019 synthetic microDiamond, and an Exradin A16 (Standard
23 Imaging, Middleton WI, USA) microchamber. Measurements were performed at 785 mm SSD and 15
24 mm depth in water and were normalized to the 60 mm fixed collimator. PDD and OAR measurements
25 were made in the same setup at multiple depths using the 60023 and 60019 detectors. The 60019
26 microDiamond was selected for this comparison since it has been previously shown to require
27
28
29
an
significantly smaller corrections for PDD and OAR measurements (generally < 1%) across this range
of field sizes than any diode or microchamber (Francescon et al. 2017). All detectors were mounted
with their stems parallel to the beam axis. Further details of the measurement technique are in
30 Francescon et al. (2017). Note that all field sizes are defined at 800 mm distance from the source.
31
32
M
33 3. Results
34
35 Tables 2 and 3 show the kΩ(0, 15 mm, f, 0, 15 mm, 60 mm fixed collimator) for the microSilicon
36
37 detector. In this situation the reference is the machine specific reference field for CyberKnife, and kΩ is
𝑓 ,𝑓
38 identical to 𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
as defined in (Alfonso et al., 2008, 2017). Figure 1 shows small field OF
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟
39 𝑓 ,𝑓
measurements using multiple detectors before and after correction using 𝑘𝑄𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟 from this study for
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
d

40
41 the microSilicon or from (Francescon et al., 2017) for the others. Before correction the microSilicon
42 measurement ratios are larger than the A16 microchamber in small fields (up to 9.4% at 5 mm field
43
pte

44 size) and smaller than the 60018 diode (up to 3.8%). After correction, the microSilicon OF’s differ from
45 the 60018 by ≤ 0.9% at all field sizes, and from the A16 by ≤ 1.6%. The microSilicon measurements
46 are most similar to those of the 60019 microDiamond (raw measurement ratios agree to < 1.8% at all
47 field sizes, decreasing to ≤ 1.0% after corrections are applied). At all fields ≥ 7.5 mm the raw
48
microSilicon and microDiamond OF measurements agree to ≤ ± 0.7% which is within the combined
49
50 uncertainty.
51
ce

𝑓 ,𝑓
52 Table 4 shows the perturbation factors contributing to 𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟
for the microSilicon detector as a
53 function of field size. All of the perturbations remain constant with field size to ≤2%. The overall
54 𝑓 ,𝑓
,𝑄𝑚𝑠𝑟 correction is mainly influenced by the encapsulating material (Pwall1) and silicon surrounding
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
55 𝑘𝑄𝑐𝑙𝑖𝑛
56 the sensitive volume (Pwall2) which drive the measurement overestimation at small field sizes. These
Ac

57
effects are partly offset at the very smallest field size by volume averaging (Pgeom).
58
59
60
Page 5 of 10 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3
4
5
6
7 Table 2. Monte Carlo calculated kΩ for fixed circular collimator OF measurement correction, which
8 𝑓 ,𝑓

pt
9 correspond to 𝑘𝑄𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟 (Alfonso et al.2017), with the PTW 60023 microSilicon detector and the
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟

10 CyberKnife M6 System. The estimated total uncertainty in each factor is < 0.75% (k=1)
11
12 Detector 𝑓 ,𝑓
13
𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟

cri
14
15
16 Field size (mm)
17
18 5 7.5 10 12.5 15 20
19 PTW 60023 0.981 0.981 0.986 0.993 0.996 1.000
20

us
21
22
23
24
Table 3. Monte Carlo calculated kΩ for MLC OF measurement correction, which correspond to
25 𝑓 ,𝑓
,𝑄𝑚𝑠𝑟 (Alfonso et al.2017), with the PTW 60023 microSilicon detector and the CyberKnife M6
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
26 𝑘𝑄𝑐𝑙𝑖𝑛
27
28
29
an
System. The estimated total uncertainty in each factor is < 0.75 % (k=1)

Detector 𝑓 ,𝑓
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑘𝑄𝑐𝑙𝑖𝑛
30 ,𝑄𝑚𝑠𝑟
31
32
M
33 Field size (mm)
34
35 7.6 x 7.7 15.4 x 15.4 23.0 x 23.1 53.8 x 53.9 84.6 x 84.7 115.0 x 100.1
36 PTW 60023 0.983 0.998 0.999 1.000 0.999 0.994
37
38
39
d

40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3
4
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16
17
18
19
20

us
21
22
23
24
25
26
27
28
29
an
30 𝑓 ,𝑓
31
Figure 1. Measured output factors for the smallest CyberKnife System field sizes before (OFdet clin 𝑚𝑠𝑟
,
𝑓clin ,𝑓𝑚𝑠𝑟
32 abbreviated to OF in the axis title) and after (𝛺𝑄𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟 , abbreviated to Ω in the axis title) correction
M
33 𝑓 ,𝑓
using the 𝑘𝑄𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟 data from tables 2 and 3 for microSilicon and from Francescon et al. (2017) for the
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
34
35 other detectors. Data is shown for (a) fixed collimators and (b) MLC. Estimated total uncertainty in the
𝑓clin ,𝑓𝑚𝑠𝑟
36 measured OFdet data is < 0.55 % (k=1). Note the change in Y-scale between (a) and (b).
37
38
39
Table 4. microSilicon perturbation factors at 15 mm depth on central axis for fixed circular collimators,
d

40
41 normalized to the same perturbation factor at 60mm field size.
42
43 ̅⁄ )𝒘
pte

Field size 𝑷𝒘𝒂𝒍𝒍𝟏 𝑷𝒘𝒂𝒍𝒍𝟐 𝑷𝒇𝒍 (𝑳 𝑷𝝆 𝑷𝒈𝒆𝒐𝒎

44 (mm) 𝝆 𝒅𝒆𝒕
45 5 0.991 0.980 0.999 0.997 1.013
46 7.5 0.989 0.990 1.001 0.998 1.002
47
10 0.991 0.994 1.001 0.999 1.000
48
12.5 0.993 0.997 1.002 0.999 1.002
49
15 0.994 0.999 1.002 0.999 1.001
50
51 20 0.997 1.001 1.002 1.00 0.999
ce

52
53
54 Figure 2 shows the difference between microSilicon and microDiamond PDD measurements over the
55 full range of field sizes. For each field size the PDD were normalized at the depth of maximum dose
56
determined using the microDiamond. At more than 10 mm depth the two measurements agree within
Ac

57
58 1% (normalized to dmax dose) for all field sizes. Closer to the surface the microSilicon detector gives a
59 significantly lower measurement than the microDiamond in all field sizes, by up to 6%-8% at the water
60 surface.
Page 7 of 10 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3
4
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16
17
18
19
20

us
21 Figure 2. Percentage Depth Dose measurements with the microSilicon detector minus measurements
22 using the PTW 60019 synthetic microDiamond detector at 785 mm SSD for the 5 mm and 60 mm fixed
23 collimators and the maximum 115 mm x 100 mm MLC field size. For each field size the PDD were
24
25 normalized at the depth of maximum dose determined using the microDiamond. Differences are <1%
26 except within 10 mm of the water surface, where the microSilicon gives a significantly lower
27
28
29
measurement.
an
Figure 3 shows kΩ(r,z,f,0,z,f) for OAR measurement. The corrections are <1% inside the field and
30 immediately beyond the field edge. At larger off-axis distances the response is more complex. For small
31
fields the microSilicon underestimates OAR and for larger fields it overestimates. The impact of this
32
M
33 effect is shown in figure 4, which compares OAR measured at large field sizes and depths, where out-
34 of-field OAR is largest, using the microSilicon, microDiamond, and 60018 diode detectors. Inside the
35 beam and just outside the field edge all agree well so, for example, measurements of FWHM are
36 consistent. Outside of the beam the 60023 measured OAR is typically 1.3% higher than the 60019
37
38
(normalised to central axis dose) at 60 mm field size, increasing to 2.3% at 115 mm x 100 mm. The
39 60023 and 60018 measurements are very similar at all positions.
d

40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 Figure 3. MC calculated correction factors for OAR measurement kΩ(r,z=100 mm,f,0,z=100 mm,f),
4
simplified to kΩ in the figure, as a function of off-axis distance for the microSilicon detector at 785 mm
5
6 SSD. Data is shown for the 5 mm and 60 mm fixed collimators at 100 mm depth. The x-axis shows the
7 off-axis distance, r, as a proportion of the geometric field radius, r0.
8

pt
9
10
11
12
13

cri
14
15
16
17
18
19
20

us
21
22
23
24
25
26
27
28
29
an
30
31
32
M
33
34
35
36
37
38
39 Figure 4. Measured OAR for the 60 mm fixed collimator and the 115 mm x 100 mm MLC field at 785
d

40
mm SSD and 300 mm depth. PTW 60023 microSilicon detector measurements are higher than PTW
41
42 60019 synthetic microDiamond measurements outside the beam by about 1.3% at 60 mm field size, and
43 by 2.3% at 115 mm x 100 mm (normalized to central axis dose). The microSilicon measurements are
pte

44 very similar to the PTW 60018 diode measurements at all positions.

45
46
47
48 4. Discussion
49
50 The purpose of this study was to evaluate the small field dosimetry characteristics of the microSiliocon
51
ce

detector, and specifically to calculate and verify the correction factors required for the TRS-483
52
53 formalism (Alfonso et al. 2017) with the CyberKnife system. The microSilicon is shown to overestimate
54 output factors at field sizes smaller than 20 mm, to a degree that increases with decreasing field size.
55 This trend is consistent with other stereotactic diode detectors. However, the correction factor
56 magnitude is significantly smaller with the microSilicon. The largest measurement correction observed
Ac

57 𝑓 ,𝑓
58 was about 2% (𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟
= 0.981 at 5 mm field size), which compares with corrections of 5% - 6% for
59 the same conditions with other stereotactic diodes (Francescon et al.2017). The microSilicon
60
Page 9 of 10 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 corrections are smaller (i.e. closer to unity) than those for the 60018 diode (Francescon et al. 2017) by
4 𝑓 ,𝑓
5 1% - 4% at field sizes ≤ 12. 5 mm. These 𝑘𝑄𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟 were confirmed by detector measurement
𝑐𝑙𝑖𝑛 𝑚𝑠𝑟

6 comparison (figure 1).

7
8 Because of these significant correction factor differences between the microSilicon and other diode

pt
9 𝑓 ,𝑓
10 designs we recommend that 𝑘𝑄𝑐𝑙𝑖𝑛 𝑚𝑠𝑟
𝑐𝑙𝑖𝑛 ,𝑄𝑚𝑠𝑟
specific to the microSilicon detector is used (i.e. for CyberKnife
11 measurements we recommend using the correction factors from this study rather than Francescon et al.
12 2017 or CyberKnife data in TRS-483).
13

cri
14 Schönfeld et al. (2019) have recently evaluated microSilicon output factor corrections for a Siemens
15
16
Artiste 6MV beam at 100 mm depth and square fields down to 5 mm x 5 mm, by measurement
17 comparison with a point scintillation detector. That work also showed that the microSilicon
18 overestimates OF at field sizes < 20 mm, but to a much lesser degree than the unshielded PTW 60017
19 diode (their largest correction was 4% with the 60023 vs. 7% with the 60017).
20

us
21 The PDD measurement comparison shows that the microSilicon is suitable for relative dose
22
measurements along the central axis over the full range of CyberKnife fields without correction.
23
24 Differences were observed between microSilicon and microDiamond detector measurements at depths
25 <10 mm, rising to 6% - 8% at zero depth. Stereotactic diodes have been previously shown to
26 underestimate close to the surface, for example with the 5 mm fixed collimator the 60018 diode is
27
28
29
an
associated with a depth-dose correction factor at zero depth of approximately 1.06 (Francescon et al.
2014b). In addition the impact of experimental set-up uncertainties and inter-detector construction
30 uncertainties is maximised in the very steep dose gradients near the surface. One repeat measurement
31 on a separate occasion and using a different microSilicon detector found the zero depth PDD difference
32 was decreased to 5%. These differences merit further investigation, but we do not consider them to
M
33 affect the suitability of the detector for commissioning measurements.
34
35 For OAR measurements the microSilicon detector underestimates OAR in the low dose profile tail (i.e.
36
37 outside the beam edge) at small field sizes and overestimates in this same region for larger field sizes,
38 which is consistent with the behaviour of other unshielded diode detectors. A user should be aware of
39 this when evaluating a beam model fit in the profile tail, and preference might be given to the synthetic
d

40 microDiamond detector, if available, for OAR measurements.

41
42 On a practical note, the microSilicon detector produces a lower signal than the 60018 diode (about 10x
43
pte

lower) consistent with the smaller sensitive volume, although this is still 10x larger than the synthetic
44
45 microDiamond. Other important detector characteristics such as response linearity and sensitivity to
46 dose-per-pulse, temperature, and radiation history are not evaluated in this work. Some of these have
47 been investigated recently by Schönfeld et al. (2019). With that qualification we conclude that the
48 microSilicon detector is suitable for small field dosimetry with CyberKnife. It requires smaller
49
50 corrections for small field Output Factor measurements than the 60018 diode which it replaces.
51
ce

52
53
54 Acknowledgements
55
56 The authors wish to thank Dr Jan Wuerfel and Dr Daniela Poppinga from PTW for loaning the 60023
Ac

57 microSilicon detector used in this study, helpful discussions on the new detector design, and their
58 comments on this manuscript.
59
60
 AUTHOR SUBMITTED MANUSCRIPT - PMB-109365.R1

1
2
3 References
4
5 Alfonso R et al 2008 A new formalism for reference dosimetry of small and non-standard fields Med.
6 Phys. 35 5179-5186
7
8 Alfonso R et al 2017 Dosimetry of small static fields used in external beam radiotherapy: An

pt
9 international code of practice for reference and relative dose determination. IAEA Technical Report
10 Series no 483, IAEA,
11
12 Asmerom G A et al 2016 The design and physical characterization of a multileaf collimator for robotic
13 radiosurgery Biomedical Physics and Engineering Express 2 017003

cri
14 Aspradakis M M, Byrne J P, Palmans H, Conway J, Rosser K, Warrington A P and Duane S 2010 Small
15 field MV photon dosimetry, IPEM Report 103, IPEM, York, UK
16
17 Bouchard H, Seuntjens J, Carrier J-F and Kawrakow I 2009 Ionization chamber gradient effects in
18 nonstandard beam configurations Med Phys. 36, 4654
19
Francescon P, Kilby W, Satariano N, Cora S 2012 Monte Carlo simulated correction factors for machine
20
specific reference field dose calibration and output factor measurement using fixed and iris collimators

us
21
22
on the CyberKnife system Phys. Med. Biol. 57 3741-3758
23 Francescon P, Kilby W and Satariano N 2014a Monte Carlo simulated correction factors for output
24 factor measurement with the CyberKnife system – results for new detectors and correction factor
25 dependence on measurement distance and detector orientation Phys Med Biol 59 N11-N17
26
27
28
29
an
Francescon P, Beddar S, Satariano N and Das I J 2014b Variation of kQclin,Qmsr(fclin,fmsr) for the
small-field dosimetric parameters percentage depth dose, tissue-maximum ratio, and off-axis ratio Med
Phys 41 101708
30
Francescon P, Kilby W, Noll JM, Masi L, Satariano N and Russo S 2017 Monte Carlo simulated
31
corrections for beam commissioning measurements with circular and MLC shaped fields on the
32
CyberKnife M6 system: a multidetector study including diode, microchamber, point scintillator, and
M
33
34 synthetic microdiamond detectors Phys Med Biol62 1076-1095 (with corrigendum 62 1642-1643)
35 Poppinga D, Kranzer R, Ulrichs A-B, Delfs B, Giesen U, Langer F, Poppe B, and Looe H K 2019
36 Three-dimensional characterization of the active volumes of PTW microDiamond, microSilicon, and
37 Diode E dosimetry detectors using a proton microbeam Med Phys 46 4241-4245
38
39 Schönfeld A-B, Poppinga D, Kranzer R, Leon De Wilde R, Willborn K, Poppe B, and Khee Looe H
2019 Technical Note: Characterization of the new microSilicon detector Med Phys 46 4257-4262
d

40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60