Evaluation of the PTW microDiamond in edge‐on orientation for dosimetry in small fields

Abstract Purpose The PTW microDiamond has an enhanced spatial resolution when operated in an edge‐on orientation but is not typically utilized in this orientation due to the specifications of the IAEA TRS‐483 code of practice for small field dosimetry. In this work the suitability of an edge‐on orientation and advantages over the recommended face‐on orientation will be presented. Methods The PTW microDiamond in both orientations was compared on a Varian TrueBeam linac for: machine output factor (OF), percentage depth dose (PDD), and beam profile measurements from 10 × 10 cm2 to a 0.5 × 0.5 cm2 field size for 6X and 6FFF beam energies in a water tank. A quantification of the stem effect was performed in edge‐on orientation along with tissue to phantom ratio (TPR) measurements. An extensive angular dependence study for the two orientations was also undertaken within two custom PMMA plastic cylindrical phantoms. Results The OF of the PTW microDiamond in both orientations agrees within 1% down to the 2 × 2 cm2 field size. The edge‐on orientation overresponds in the build‐up region but provides improved penumbra and has a maximum observed stem effect of 1%. In the edge‐on orientation there is an angular independent response with a maximum of 2% variation down to a 2 × 2 cm2 field. The PTW microDiamond in edge‐on orientation for TPR measurements agreed to the CC01 ionization chamber within 1% for all field sizes. Conclusions The microDiamond was shown to be suitable for small field dosimetry when operated in edge‐on orientation. When edge‐on, a significantly reduced angular dependence is observed with no significant stem effect, making it a more versatile QA instrument for rotational delivery techniques.


| INTRODUCTION
Quality assurance (QA) for small field radiotherapy is a challenging task requiring new detectors and QA methodologies. Codes of Practice (CoP) for conventional external photon beam radiotherapy are not suitable for small field dosimetry as they do not account for the lack of lateral charged particle equilibrium (LCPE) or occlusion effects. 1 The IAEA TRS-483 CoP for small field dosimetry recommends detectors used for QA should be: small relative to the minimum field size and the range of the secondary charged particles, have a high signal to noise ratio (SNR), high spatial resolution and also be energy, dose rate, and angular independent in response. Volume averaging effects can be avoided by the use of detectors with submillimeter spatial resolution, such as the PTW microDiamond, 2-5 IBA razor diode detector, 6 and the edgeless silicon diode, 7 however, these detectors do not address some of the more serious perturbation effects. A tissue equivalent, small volume detector can be made for plastic scintillators but these detectors suffer from a large temperature and humidity dependence as well as nonlinearity at low doses. 8 In other detectors, density-based perturbation effects are caused by the inhomogeneity of the detector volume and packaging with respect to the surrounding medium. Perturbation is created due to the mismatch in stopping power ratios of the detector and its packaging materials relative to water which can lead to large variations in the detector response. Alfonso et al, presented a methodology where the detector response variation with field size can be corrected for by using a detector-specific sensitivity correction factor, however, this assumes a certain detector orientation, angular independent response, and isocentric delivery. 9 Diamond is a natural candidate for small field dosimetric applications given its tissue equivalence, 10-12 radiation hardness, 13,14 and near energy independent response to water for x rays. 15 The uptake of diamond-based devices has been hampered by its lower sensitivity which can be quantified in the energy required to create an electron hole pair (E e/h = 13 eV 16 ). Furthermore, the density of diamond (ρ = 3.52 g cm −3 ) will increase perturbation effects for small field dosimetry QA. The PTW microDiamond (PTW 60019, PTW, Freiburg, Germany) is currently one of the most prolific diamond-based detector in clinical use in radiotherapy. The microDiamond detector developed by the University of Rome Tor Vergata 17,18 and commercialized by PTW, utilizes synthetic single crystal diamond featuring a metal/intrinsic diamond/p-type diamond (m-i-p+) structure. The result of the m-i-p + structure is a built-in potential allowing for the device to run in passive mode, that is, zero applied bias. There exists within the literature an ongoing debate regarding the appropriateness of the PTW microDiamond for small field dosimetry with conflicting reports of over response, 2,3 water equivalence, [19][20][21][22] and under response for fields <1 cm 2 . Recent work has also quantified an additional effect of radiation-induced charge imbalance, in the PTW microDiamond. 23 At field sizes of length <2 cm radiation induced charge in the electrical contact of the PTW microDiamond is reported to result in an overresponse of the device. Despite these limitations for dosimetry applications the microDiamond remains an area of interest for QA with its response having recently been characterized for use in an MRI linac 24 and for the MRI-associated surface dose from electron contamination. 25 One of the key advantages of the PTW microDiamond is its high spatial resolution when operating in edge-on orientation, 21,26,27 however, the IAEA TRS-483 CoP requirement of a face-on orientation for all QA measurements means that this micron scale spatial resolution is unrealized. The edge-on and face-on orientation are referred to by the terms perpendicular and parallel orientation in the CoP, respectively. This edge-on (perpendicular) orientation has not previously been characterized in the context of small field dosimetry. Furthermore, diamond detectors have been investigated for both stereotactic ablative radiotherapy (SABR) and volumetric arc therapy (VMAT), 28,29 making the angular dependence of this orientation of great interest. In this study, the PTW microDiamond is characterized in both edge-on and face-on orientations for the first time. Additionally, the angular dependence of the PTW microDiamond is investigated thoroughly in order to evaluate the potential use in edge-on orientation.

2.A | Percentage depth dose measurements
In order to assess the appropriateness of the microDiamond for small field the PDD measurements were performed for 1 × 1 and  A list of detectors used in this study and their corresponding sensitive volumes is presented in Table 1.

2.B | Lateral beam profiles
To assess the impact of orientation on the spatial resolution of the microDiamond for small field QA, field profiles were performed for 0.5 × 0.5, 1 × 1, and 3 × 3 cm 2 fields in face-on and edge-on orientations at 10 cm depth. The profiles were performed at 100 cm SSD with jaw defined field for in-plane and cross-plane profiles, with multiple measurements taken at each point and averaged. The field sizes are defined for the beam at an SSD of 100 cm.

2.E | Stem effect
The

2.F | TPR 20,10
Tissue to Phantom Ratios (TPR) were obtained for the microDiamond in edge-on mode as well as the IBA Razor diode and compared with the CC01 ionization Chamber. In this work the TPR 20,10 (S) is defined as the ratio of detector response R at 20 cm depth to 10 cm, R 20 /R 10 .
S denotes square field size ranging between 0.5 × 0.5 cm 2 and 10 × 10 cm 2 . A source to detector distance (SDD) of 100 cm was used with 30 × 30 cm 2 blocks of SW which was used to make the relevant build-up depth and 10 cm of back scatter material.  Table 2. The values of d max for the four data sets
The full width half maximum (FWHM) and average penumbra width for in-plane and cross-plane measurements for 0.5 × 0.5, 1 × 1, and 3 × 3 cm 2 field are presented in Table 3. The penumbra is consid- Similarly, penumbra widths are between 5.6 and 8.3% narrower when measured in edge-on orientation. The face-on orientation measures a larger or equal penumbra for all field sizes compared to the edge-on orientation and is also larger or equal to measurements taken with the Razor diode from 1 × 1 cm 2 . The thinned sensitive volume of the edge-on orientation produces a reduced volume averaging effect31 and is thus more appropriate for this measurement.

3.C | Output factor
The photon output factor (OF) is shown in Fig. 7 for square field sizes from 0.5 × 0.5 cm 2 up to 8 × 8 cm 2 and are normalized relative to 10 × 10 cm 2 field. Percentage difference graphs of the OF are shown in Fig. 8 reporting that the microDiamond in both orientations agrees within 1% down to the 2 × 2 cm 2 field size for both The IAEA TRS-483 CoP provides correction factors for the microDiamond and CC01 and these have been applied, where the data were available in Fig. 8. The data presented from a study by Poppinga et al. 6 suggest that no significant correction factor is required to be applied to the Razor chamber until the field size is smaller than 1 cm 2 and is therefore used as the reference chamber in Fig. 8  The variation in the response for face on-orientation is reported in Table 4 with the data between angles 70°and 120°omitted. For this subset of angles, the microDiamond in face-on orientation still has a significant angular dependence with the range of response as large as 12% at the 10 × 10 cm 2 field. In face-on orientation, the microDiamond is therefore highly angular dependent and would be inappropriate for measurements that involved multidirectional beam geometries.

3.D | Angular dependence
In the edge-on phantom, the range in angular response through 240°reported in Table 5 shows only a 2% variation for the 2 × 2 cm 2 and 3 × 3 cm 2 fields. This increases up to 28% for a 0.6 × 0.6 cm 2 field and is likely to be at least partially related to the introduction of jaw sag and/or a slight error in the vertical alignment of the detector which is only prevalent at very small fields. No fluence monitor chamber was used to correct for variations in machine output so this may also account for small variations. The main advantage of using the microDiamond in edge-on mode is therefore an almost angular independent response even for small fields of 2 × 2 cm 2 . In Fig. 9(a) it is observed that smallest deviation in angular response is around the 0°position where the orientation has transitioned to edge-on. In edge-on mode, the microDiamond is therefore also insensitive to detector tilt.

3.E | Stem effect
The values of the stem effect for the microDiamond are presented in Table 6 with the largest observed stem effect of 1.00% for the 1 × 3 cm 2 field size. By the 3 × 10 cm 2 field size, the stem effect is negligible for both energies recording less than a 0.2% increase in OF when the beam runs parallel to the cable. The close agreement of the stem effect between the 1 × 10 cm 2 and the 1 × 3 cm 2 suggests that the increased signal is primarily coming from interactions around the high Z electrodes of the microDiamond and not the cable.

3.F | TPR 20,10
The measurements of the TPR 20,10 (S) are presented in Table 7.The microDiamond has less than a 1% discrepancy to the CC01 at all field sizes. The Razor diode had a 2.61% difference to the CC01 for the 0.5 × 0.5 cm 2 field and a difference of <1% at all larger field sizes. The IAEA TRS-483 CoP suggests the use of ionization chambers for beam quality measurements due to the small mismatch in stopping power ratios between air and water, but also considers that the size of the detector should not perturb the field. The data presented in Table 7 show that the TPR does vary with field size and that the collimation of small field with MLCs will impact on beam quality. The close agreement between the microDiamond and the CC01 identifies the potential for microDiamond in edge-on mode to   new rotation-based systems, such as for MRI linacs. 33 The angular dependence in edge-on orientation is, however, not expected to be the same within a magnetic field. Inside the MRI linac the secondary electrons produced in the surrounding material are affected by the Lorentz Force depending on the field orientation. 24,25 The isolation and quantification of density perturbation effects on orientation will also be investigated in relation to the impact of magnetic field in future work.

| CONCLUSION
The first characterization of the PTW microDiamond in an edge-on orientation has been undertaken, demonstrating its advantages for dosimetry in small fields. The microDiamond in an edge-on orientation has an angular independent response down to a 2 × 2 cm 2 field with a maximum deviation in angular response of 2%. For a 0.5 × 0.5 cm 2 field when edge-on, the microDiamond on average measures the FWHM and penumbra between 1.1-4.4% and 5.6-8.3% narrower respectively, compared to the recommended face-on orientation. The stem effect introduced when using the detector in edge-on orientation was at a maximum producing a 1% increase in response. Both orientations require correction factors when taking measurements of field sizes <2 × 2 cm 2 due to an observed overresponse. Correction factors for the edge-on orientation will be the focus of future work. This study has demonstrated the advantages and potential versatility of using the microDiamond in edge-on orientation for applications outside small field OF measurement.

ACKNOWLEDGMENT
The authors acknowledge the support of the NHMRC (grant APP1093256) and AINSE (grant RPP10102) and thank the team of the Australian Synchrotron Imaging and Medical Beam line and Illawarra Cancer Care Centre for their assistance. This research has been conducted with the support of the Australian Government Research Training Program Scholarship.

CONFLI CT OF INTEREST
No conflict of interest.