COMP report: CPQR technical quality control guidelines for major dosimetry equipment

Abstract The Canadian Organization of Medical Physicists (COMP), in close partnership with the Canadian Partnership for Quality Radiotherapy (CPQR) has developed a series of Technical Quality Control (TQC) guidelines for radiation treatment equipment. These guidelines outline the performance objectives that equipment should meet to ensure an acceptable level of radiation treatment quality. The TQC guidelines have been rigorously reviewed and field tested in a variety of Canadian radiation treatment facilities. The development process enables rapid review and update to keep the guidelines current with changes in technology (the most update version of this guideline can be found on the CPQR website). This article provides guidelines for quality control testing of major dosimetry equipment.

Suggested methods for measurement of ion collection efficiency and polarity correction may be found in AAPM TG-51. 3 For flattening filter free (FFF) beams, the effects of higher dose rates should be investigated, as recommended by AAPM TG-51 Addendum. 4 Leakage tolerance and action levels are based on the ratio of leakage versus ionization current/charge. Since collection potential (voltage) is difficult to accurately measure with the chamber connected, the user may rely on the internal device readout for the measurement of the collection potential reproducibility test (ISS6). ESS1, BSS1: Based on AAPM TG-40. 12 IFS1-IFS5: Tolerances based on AAPM TG-40. 12 Suggested methods for measurement may be found in AAPM TG-51. 3 IFS6: Based on clinical experience. AFS1, AFS2: Based on clinical experience and AAPM TG-40. 12 Since collection potential (voltage) is difficult to accurately measure with the chamber connected, the user may rely on the internal device readout for the measurement of the collection potential reproducibility test. AFS3: Modified frequency from AAPM TG-40 12 based on clinical experience. ECC1: Prior to their use, detectors, cables, and connectors should be checked for any defects and for functionality. An unusually high leakage level or lack of reproducibility of measurements is an indication of a problem and would need to be addressed.

2.B | Detectors for non-reference dosimetry
These are detectors used to measure dose from a radiation source as a method of ensuring the stability of the device on a routine basis. They can also be used to determine the absolute dose in a phantom or received by a patient following a cross-calibration process. Some of these devices in use include ionization chambers, diodes, thermoluminescent dosimeters (TLDs), metal-oxide semiconductor field-effect transistors (MOSFETs), optically stimulated luminescence (OSL) systems, scintillating fibre dosimeters, radiographic films, 7 and radiochromic films. 8 Both types of films are integral parts of routine quality assurance for intensity-modulated radiation therapy (IMRT) treatment plans and for stereotactic radiosurgery.

2.C | Basic measurement devices
Most secondary and field standards are vented ionization chambers and as such, are subject to local atmospheric conditions. Therefore, thermometers, barometers, and hygrometers will be used during reference dosimetry measurements. Basic distance checks will be achieved with a quality ruler or caliper. A quality stopwatch will be used for accurate time measurement. Spirit levels (with or without digital angle display) could be used for levelling scanning water tanks and other measurement phantoms or devices. A self-adjusting laser system projecting two perpendicular laser lines may be used to check the horizontality and verticality of room lasers.

2.D | Automated beam scanning devices
Automatic remotely controlled water scanners comprise a water tank and a mechanism for holding and moving a radiation detector through the beam. They range in sophistication from ion chamber motion/measurements along a single vertical axis (1D water tanks) to a motion along two (2D water tank) and three directions (3D  12 Investigation of linearity and supralinearity for a sample of a few TLDs from a batch. ERD1: Based on AAPM TG-40. 12 Multiple TLDs can be cross-calibrated simultaneously against an ion chamber measured dose at a reference depth in a solid phantom using a uniform radiation field. IRD2: Can be established using classic H&D curve for one film for each new batch. Effects of batch film changes should be routinely assessed. Various techniques for obtaining a sensitometric and a dose-response curve are described in AAPM TG-69 7 for radiographic films and in AAPM TG-55 8 for radiochromic films. WRD1: Testing to follow manufacturer recommendations. ARD1: Based on AAPM TG-69 7 for radiographic films and on AAPM TG-55 8 for radiochromic films. IRD3, IRD4, ARD2: Based loosely on AAPM TG-40 12 and clinical experience. IRD5, IRD6: Based on AAPM TG-40. 12 ARD3: Based on AAPM TG-40. 12 Absolute dose calibration to be done if required. IRD7, IRD8: Energy dependence of MOSFETs can be addressed by performing an absolute dose cross-calibration in the beam energy and conditions they are intended to be used. 13 Cross-calibration for each beam quality against an ion chamber dose, as per AAPM TG-51 3 or TG-43, 14 following manufacturers' recommendations. ARD4: Absolute dose cross-calibration in the beam energy and under conditions they are intended to be used. IRD9: Linearity of the OSL detectors should be checked prior to use to assess the dose range at which the dosimeter remains linear. IRD10, ARD5: Commercially available OSL detectors show minimal energy dependence in the megavoltage clinical energy range 6À25 MeV. Substantial energy dependence has been found in the kV range. Therefore, the same absolute calibration factor can be used in the megavoltage energy range, while an energy-dependent calibration should be done for energies in the kV range. IRD11: Linearity of the scintillating fibre dosimeter (SFD) should be checked prior to use to assess the dose range at which the dosimeter remains linear. IRD12, ARD6: Commercially available SFDs show minimal energy dependence in the megavoltage clinical energy range 6À20 MeV. Substantial energy dependence has been found in the kV range. Therefore, the same absolute calibration factor can be used in the megavoltage energy range, while an energy-dependent calibration should be done for energies in the kV range. IRD13: The signal from plastic scintillators contains Cherenkov radiation generated in the light guide, which results in an undesired stem effect. A stem removal technique needs to be implemented to keep this effect below stated specifications.
| 21 etc.) on measured data, and for converting the ionization depth curves into dose according to various protocols. 3,9 Also available are smaller 3D scanning water tanks that fit into the gantry bore of tomotherapy units or that are adapted specifically for tissue-phantom ratio (TPR) type measurements of stereotactic fields; these are subject to the same quality control tests as larger scanning water tanks.

2.E | Machine quality assurance devices
Megavoltage beam parameters such as output, field size, flatness, symmetry, beam energy, and constancy can be measured on a routine basis with a variety of devices which are more convenient to use than the water scanner. These devices may consist of one or more two-dimensional detector arrays of diodes or ionization chambers and may have software for processing, analyzing, and tracking measured data. These devices, which consist essentially of twodimensional detector arrays, are easy to setup and use, and their multi-detector construction involving ion chamber and/or diodes makes them useful in the monitoring of technologies such as dynamic wedge and IMRT beam quality assurance. 10 12 In addition, the manufacturers' acceptance test procedures may be used to modify the user's criteria. IMQ6: Based on clinical experience and manufacturer's recommendations. If devices are used across a range of beam energies, care must be taken to investigate their energy dependence and ensure that the appropriate calibration factors are applied for each measurement. AMQ1: Based on clinical experience and manufacturer's recommendations. Array calibration ensures that all detectors in the array have the same sensitivity and thus eliminates response differences between individual detectors of the array. The resulting calibration factors may be energy-dependent. Array calibration procedures and protocols are device-specific and are provided by all vendors. Recalibration intervals depend on the type of detectors in the array (ion chamber or diode) and on the clinical workload. Vendor's guideline for array recalibration intervals can be followed.
T A B L E 6 Treatment delivery quality assurance devices.

2.G | Phantom materials
While water is the reference phantom material for clinical reference dosimetry, solid phantoms are typically used for routine measurement. These devices may have radiation absorption properties and interaction coefficients similar to water, and may also be available in other materials such as acrylic, bone, lung, or muscle. The phantom may have "slab" geometry or be anthropomorphic. Anthropomorphic or "humanoid" phantoms are often constructed so as to accommodate TLD, MOSFETs, and film measurements. Motion phantoms that incorporate various forms of detector or target movements are also available for assessing 4D imaging and treatment gating capabilities (see Tables 1-7).

ACKNOWLEDG MENTS
We would like to thank the many people who participated in the production of this guideline. Note: ITQ1: Based on clinical experience and manufacturer's recommendations. It should be possible to attach the gantry mount accessory tightly on the gantry and to fix the detector array on it so that the detector does not move as the gantry and/or collimator rotate. ITQ2, ITQ3: Based on clinical experience. With the detector array fixed on the gantry mount, the central axis of the detector array should align with the linac crosshair and the detector plane should be at isocentre. A 2 mm tolerance could be used here. Gross errors in the alignment and positioning can be corrected by adjusting the phantom setup in the TPS or by manipulation of device measurements. Also applies to relevant beam quality assurance devices. ITQ4: Based on gantry/collimator angle indicators tolerance from AAPM TG-40 12 and AAPM TG-142. 15 ITQ5, ITQ6: Based on AAPM TG-40. 12 Manufacturers' specifications can be used to set device-specific tolerance and action levels. ITQ7: Tolerances based on AAPM TG-40 12 and review of manufacturers' specifications. ITQ8: For each TPS, care must be taken to ensure that dose import parameters are setup correctly for TPS co-ordinates to match those of the measuring device. ITQ9: Same as IMQ6. ATQ1: This is a consistency check based on clinical experience: a static field and an IMRT DQA plan can be created on the CT data set of the device in the TPS. These plans are periodically delivered on the device for consistency checks and analyzed with the gamma index parameters indicated. For the case of a static field, tighter tolerances can be used. However, the passing criteria can be adjusted locally based on the accuracy of the beam model of the TPS. ATQ2: Same as AMQ1. ATQ3: Based on clinical experience. Absolute dose cross-calibration (at each beam quality) must be done following vendor's recommendations and against an ion chamber dose obtained following AAPM TG-51, 3 International Atomic Energy Agency (IAEA) TRS-398, 6 or AAPM TG-148. 16 After transfer of ion chamber dose to the device, the latter can be irradiated with the same beam used for calibration and the dose measured by the reference detector should agree with the ion chamber dose within indicated tolerance levels. This setup can also be used for routine checks of the absolute calibration of the device. Recalibration frequency is suggested by vendors and depends on workload for diode arrays. If devices are used across a range of beam energies, care must be taken to ensure that the correct calibration factors are applied. Note: IPM1, IPM2: Inspection and radiographic verification prior to use is recommended. The tolerance depends on the intended use of the material and may be appropriately chosen by the user. this manuscript has been made possible through a financial contribution from Health Canada, through the Canadian Partnership Against Cancer.

CONF LICT OF I NTEREST
The author declares no conflict of interest.