Schweizerische Gesellschaft für Strahlenbiologie und Medizinische Physik Société Suisse de Radiobiologie et de Physique Médicale Società Svizzera di Radiobiologia e di Fisica Medica Swiss Society of Radiobiology and Medical Physics

printer friendly version

RECOMMENDATIONS No 7

Webversion 12.02.1999
This  document is the HTML-version of the published version ISBN 3-908125-23-5
To obtain the published version contact the corresponding author

Members of the working group

    Dr. Ernst BORN
            Abt. für Med. Strahlenphysik, Universität Bern, Inselspital, 3010 Bern

    Antonella FOGLIATA-COZZI
            Ospedale San Giovanni, 6500 Bellinzona

    Florica IONESCU
            Rätisches Kantonsspital, Radio-Onkologie, 7000 Chur

    Dr. Victor IONESCU
            Hirschbühlweg 18, 7000 Chur

    Pierre-Alain TERCIER ^*,
           Institut de Radiophysique Appliquée, Centre universitaire, 1015 Lausanne
           new address: Service de Radio-oncologie, HFR - Hôpital Cantonal, 1708 Fribourg,

*Corresponding author

INTRODUCTION

1. COMMISSIONING
    1.1. PREREQUISITES OF COMMISSIONING
        1.1.1. Vendor Tasks
        1.1.2. User and System Administrator Tasks
        1.1.3. Preliminary Checks for Commissioning
    1.2. CHECKS FOR COMMISSIONING
        1.2.1. Non-Dosimetric and Non-Geometric Checks
        1.2.2. Geometry
        1.2.3. Devices (peripherals)
        1.2.4. Dosimetric Measurements for TPS Validation

2. TESTS AFTER IMPLEMENTATION OF A NEW SOFTWARE RELEASE
    2.1. INTRODUCTION
    2.2. ACCEPTANCE TESTS FOR A NEW SOFTWARE RELEASE NOT AFFECTING PHYSICAL ALGORITHMS
        2.2.1. Non dosimetric checks
        2.2.2. Dosimetric checks
        2.2.3. Verification of all potentialities of the TPS

3. TESTS AFTER ADDITION OF A NEW THERAPY UNIT AND/OR IMPLEMENTATION OF NEW BEAM DATA

4. REPEATED TPS CHECKS
    4.1. INTRODUCTION
    4.2. TESTS AND FREQUENCIES
        4.2.1. Peripherals
        4.2.2. Monitor Units (MU)
        4.2.3. Standard treatment plan
        4.2.4. "Checksum"

BIBLIOGRAPHY

APPENDIX

With the introduction of CT based computerised treatment planning in radio-oncology the treatment planning system (TPS) has become a key element in the radiotherapy process. Regarding patient safety and success of therapy, its accurate and stable functioning is an issue of highest importance. For these reasons quality assurance (QA) procedures have to be implemented for the commissioning phase and for the clinical routine running of a TPS.

Being a hardware and software system, each TPS may incorporate errors [14] from any of these different kinds:

• Software implementation errors may result in a system behaviour which differs from system specifications.
• Inaccuracies in algorithms and models could produce dose results not accurate enough to be used in clinical routine.
• Errors could result from inaccurate or wrong model parameters being applied by the operator ¹.
• Errors could result from technical deficiencies of computer peripherals (printers, plotters, etc.).
• Intentional or unintentional changes in data or software files could produce erroneous results.
• Errors of different kinds could be introduced in the system during the software upgrade process.

This recommendation is the counterpart of the issue of the SSRMP "Physical and Dosimetric Checks in Teletherapy" [27]. Its purposes are defined in the same way:

• to provide help to the medical physicist for the implementation of quality assurance procedures for TPS for external therapy with photons and electrons.
• to be used during evaluation and commissioning of a TPS, in case of implementation of new beam data or new software releases and also during routine clinical use.
• to create an unity of doctrine in QA of TPS in Switzerland
• to establish a base document for the public health authority.

It is the medical physicist's responsibility to implement the QA procedures described in this document

Chapter 1 describes QA Procedures (QAP) for commissioning a TPS for both photons and electrons beams, chapter 2 and 3 deal with QA aspects after implementation of new software releases and new beam data respectively and chapter 4 lists QA recommendations for routine use of a TPS. After a bibliography of related literature, the appendix contains a summary of tolerances and frequencies of tests.

Future publications of QA guidelines by other national or international authorities, may necessitate a revision of this document.

Commissioning

Commissioning of a TPS by a qualified medical physicist is necessary before the system is used for clinical purposes. Commissioning has to validate the proper functioning of the system according to the vendor's specifications and to clinical requirements. The results can be used to establish standards of acceptance for the demonstration of the correct working of the TPS in regular QA-checks. Calculations for commissioning test cases shall be compared against measurements done at the same time as those used to characterise the treatment unit model².

1.1. Prerequisites of Commissioning

Prior to the commissioning process, the following points and suggestions should be considered.

1.1.1. Vendor³ Tasks

a) Generalities

• The IEC document [13] specifies the following vendor tasks :
• The vendors shall fulfil the conditions described in the IEC document concerning the safety of radiotherapy TPS.
• Coherence of the units
• All distances and linear dimension units shall be in centimetres.
• All values of radiation (printed or displayed) shall include their units.
• Units of radiation should conform to the SI convention.
• Password protection, or the use of a key, shall be employed to ensure that only authorised persons perform treatment planning.
• All data entered by the operator or acquired from a device or network shall be compared against fixed limits and not accepted if outside of these limits. Other consistency checks on data should also be appropriate to the expected nature of the data.
• The vendor should provide high quality training for the operators. For sophisticated 3D systems, this training should include not only teaching the operators what the effect of a particular operation inside the software will be, but also concentrate on the strategies for planning that are useful, and on various issues that only the experienced operator may come in contact with [34].
• A total protection against unauthorised access by the network is difficult to implement. The vendors shall take steps to ensure this point. They should inform the system administrator (SA, see 1.1.2.) about the method to be used as a precaution.
• The vendor should make the data of the beam library accessible. He/She should allow the SA to enter, inspect and modify all relevant data.
• The vendor shall provide a set of patient's files and beam data for demonstration purposes. The beam data can be taken from the report [1] of the "Task Group #23 of Radiation Therapy Committee".
• The vendor should provide sufficient software tools to enable the operator (and SA) to check the accuracy of the data introduced in the beam library (e.g. plots/printed tables for profiles, PDD, TMR).
• The vendor should provide a checksum utility to perform tests on the beam data library and executable files.
• In case of new releases, the vendor should provide the possibility to reuse the files of previously calculated cases. The archiving and reading back of patient files shall be possible. The re-calculation of such cases shall be and correct.

b) General requirements for planning [13]

• Planning shall not be possible using an incomplete treatment unit⁴ or patient anatomy model unless the operator overrides a cautionary message which states that the model is incomplete.
• It shall not be possible to specify beam limiting devices or beam modifiers other than those allowed at the selected treatment unit model.
• It shall not be possible to specify customised blocking devices and customised beam modifiers as being in any position relative to the beam source other than those allowed at the selected treatment unit model.
• It shall not be possible to specify a beam size or position which is not within the range of the beam limiting device or applicator at the selected treatment unit model, or larger than the maximum size for a particular beam modifier (such as a wedge).

c) Documentation

• Extensive documentation on how the TPS software works should be available, including a description of the overall design, the theory of operation, the limitations, and detailed information on what happens as each step of the planning process is performed [34]. The vendor should supply descriptions of the calculation and geometry algorithms used in their systems (see next section for details). The understanding of the algorithms allows the operator to appreciate the limitations imposed by the physical models and to design simple checks to ensure the proper functioning of the system in clinical routine (e.g. hand calculation of monitor units).
• The vendors shall provide information on the hardware and software QA program which is used to design, develop, test, document, and release the software. A detailed description is not necessary, but the vendor should supply it on request [34].
• The vendor should release detailed descriptions of the datafile formats and contents which has been used for data import and export, along with examples of correct implementation of this data transfer mechanism [34]. Whenever possible, generally accepted protocols (ACR/NEMA [2], DICOM3 [7]) should be used.

d) Algorithms

• All algorithms used for dose calculation shall be clarified in the technical description. This shall include a description of the factors taken into account by the algorithm, the mathematical equations which form the basis of the calculation, and the limits of all variables used in the equations. References to the literature shall be provided for published algorithms [13].
• Where a choice of algorithms is provided for a particular calculation, the instructions for use shall clearly identify the relative advantages and disadvantages of the different algorithms [13].
• For each algorithm used, the accompanying documentation shall state the accuracy of the algorithm relative to measured data for at least one set of pre-defined conditions [11],[13].

1.1.2. User⁵ and System Administrator Tasks

• The operator is responsible for learning how to use the TPS correctly. The user (institution) should also identify for each TPS unit one physicist to receive extra training, so that he/she can fulfil his/her responsibility as the overall supervisor of how the planning system is used at that institute [34]. This person will be called system administrator (SA) for the rest of this document.
• The commissioning shall be performed or supervised by the SA.
• The SA shall have a full understanding of the operation of the hardware and software.
• The uncertainty associated with the measured data, data entry and output shall be noted. Known inaccuracies in the planning algorithm, through publications and/or vendor documentation, should be understood by the SA.
• A log-book shall be used for the commissioning of the software and used also, after commissioning period, during routine use. The aim of this log-book is to document all the events and changes concerning the TPS. The log-book is maintained by the SA.
• The SA shall have total control of the data of the beam library; it is recommended that he/she should be able to enter, inspect, and modify, by himself, all relevant data [11].
• It is not possible to cover all possible applications but checks should cover a representative set of those applications used in the institute.
• The SA should establish standard procedures for the use of the TPS. These procedures shall be carefully tested. The operators shall follow the standard procedures.

1.1.3. Preliminary Checks for Commissioning

TPS hardware tests are required in order to ensure that both the computer and its peripherals are operating according to specifications. Most computers have an automatic system diagnostics that include tests for the processor, memory and disk operation. In addition the following items are to be tested:

• Plotter, printer, digitizer : connection and self-tests
• Display : the distortion of the video display shall be checked using a grid displayed on screen
• Presence of a checksum utility to be performed on the executable files and beam data files. This checksum utility could be very helpful to track any abnormal change in important files.

1.2. Checks for Commissioning

1.2.1. Non-Dosimetric and Non-Geometric Checks

Allowed ranges and directions of linear and rotational scales for technical treatment machine parameters should correspond to the actual situation for individual treatment units. If not, this shall be taken into account and a specific procedure shall be defined concerning the transfer of set-up information from treatment plans to the corresponding therapy unit.

The data transfer from TPS to the simulator and/or the accelerator shall be tested.

All the parameters describing beams shall be checked for completeness, availability and consistency on screen or paper (e.g. isodose chart).

All patient identification details shall be checked for completeness, availability, consistency and uniqueness. All reproductions on hard copies or printed pages shall be complete with regard to date, time, plan identification number and version number of the calculation.

An independent check shall be undertaken to ensure that the basic input data has been entered correctly. This can be done by printing out tables of entered beam characterisation parameters, e.g. values of TAR, TPR, depth doses, output factors etc., if the algorithm is based on tabulated data. The aim is to ensure correctness of data input.

1.2.2. Geometry

The accuracy and completeness of the 3D-models of the patient and the beams have to be checked (the required tolerance regarding to distances is 0.1 cm). Every TPS has it's own functions and these have to be tested according to their use. The following list shall be completed according to the TPS particularities:

a) Geometry

• The transfer of patient data from CT to TPS shall be geometrically correct.
• The transfer of images (CT, MRI, PET, ...) shall preserve the pixel values and magnification and linearity of the image.
• The (manual or automatic) definition of volumes of interest (geometrically simple objects) shall correspond to the real dimension and position and shall not be distorted in cardinal and oblique sections.
• The outline of a shaped beam cut by a cardinal or oblique section shall be correct. It could be tested by comparison with a light field on the treatment machine.
• Bolus, although often considered on a beam-by-beam basis, is sometimes treated more as a part of the patient anatomy, at least in how it interacts with dose calculation algorithms. The following tests shall be applied [34]:
• Verification that the bolus is correctly added into the calculation algorithm,
• Verification that the hardcopy output and other display of geometric information (e.g., the SSD displayed when bolus is used on the surface of the patient) agree with the stated method,
• Verification of the effect on monitor unit calculations,
• Verification that density of the bolus is correct.

b) 3D options

• Once a number of outlines are defined, the multiple outlines defined on slices (CT slices) shall then often meshed into a 3D surface description if they are to be used for a 3D model of the patient (this can also apply to a 3D volumetric representation). The algorithm used to create this 3D surface can be quite complex [34]. Therefore, the QA program for this feature shall consist of :
• A straightforward testing of a set of extreme cases which could be mishandled (square outline next to a three sided outline, use of sharply pointed outlines, outlines with one or two points only, unclosed outlines, outlines missing on adjacent slices, etc.),
• A more rigorous testing of a subset of manually entered cases (sphere, cylinder, cube or pyramid) where the exact formation of the surface mesh is easy to estimate and can be compared to the one calculated by the TPS,
• A testing of interpolated outlines.
• If it is possible to extract or "cut" an outline from that 3D description. This feature requires additional testing [34]:
• The general limitations to the implementation (e.g. the algorithm can only cut axial outlines),
• The results with complicated structure issuing in multiple independent outlines in a particular slice or cut,
• The shape of simple objects (sphere, cylinder, cube, ...) cut in different ways (cardinal or not)

Note : Not all of these options are available in all TPS.

c) Beam's-eye-view

• The relative size, position and shape of known objects in beam's-eye- and observer's-eye-view shall be correct.
• The beam's-eye-view shaping and size of the beam (blocks, multileaf collimator) shall correspond to the reality.
• The Digitally Reconstructed Radiograph (DRR) shall be compared to the original films from the simulation procedure (with phantoms where measurements of distances can be performed).

d) Dose-Volume Histogram

• The total volume on DVHs (if they exist) shall be checked and compared with other methods of computation of the size of volumes of interest (VOIs).
• Consistency checks with the isodoses charts shall be applied.

1.2.3. Devices (peripherals)

a) Printing/Plotting Devices

On each treatment plan printed or plotted, two scales in both directions (horizontal and vertical) with centimetre marks at least over 10 centimetres are required. However all the marks shall be verified with a ruler, they must be at the right place (i.e. no distortion and no displacement).

• Tolerance: 0.1cm

b) Digitizer

The vertical and horizontal scales are to be tested along 10 cm showing an agreement better than 0.1cm.

• Tolerance: 0.1cm

c) Film scanner

An object on a film with particular dimensions is scanned. The size displayed by the TPS is then compared to the real one.

• Tolerance: 0.1cm

d) Computed Tomography (CT)

General QAP of CT are not the subject of these recommendations, additional information on CT QAP can be found in the literature [19],[20],[25],[32]. Issues covered here are the transfer of CT data to the TPS and the use of CT data by dose calculation algorithms.

• To check CT image transfer a QAP has to be set up consisting of several CT examinations of an object of known shape and dimension. The QAP has to check the correctness of the transfer of the complete 3D information (position and magnification) from CT to TPS for all relevant patient set-ups (head-first, feet-first, prone, supine, left, right) and CT parameters (scan and reconstruction diameters, tube angle for topograms - scout views -, etc.).

Geometrical distorsions shall be less than 0.1 cm

• Hounsfield numbers of the CT images normally are used by the TPS to extract radiation absorption properties of the tissues at therapeutic beam qualities. QAPs therefore have to be set up to ensure the correct transfer and use of the Hounsfield numbers. The details of these QAPs depend on TPS algorithms. Normally CT examinations have to be made of a phantom containing tissue like material of known chemical compositions and densities. TPS extracted parameters (physical or electron densities, absorption coefficients, scattering or stopping powers, ...) can then be checked directly. Knowing the dose calculation algorithms of the TPS, the impact of inaccuracies in extracted parameters on dose calculation results can be estimated.

To give and idea of the expected accuracy :

• Electronic density relative to water (<=1.5): Tolerance is 0.05
• Electronic density relative to water (>  1.5): Tolerance is 0.1

e) Block cutting device

Information to produce blocks (and compensators) have to be transferred from the TPS to the block cutting device. The proper functioning of data transfer shall be checked by examining blocks of clinically relevant shapes.

f) Archiving and reading back of patient data

If the TPS includes an archiving system, the correctness of treatment plan parameters and patient data read back from the archive has to be checked (correct treatment plans irrespective of TPS development and addition of therapy units etc.). The scale of outlines, the shape of the beam, the date (for a Cobalt-60 unit) are to be checked.

1.2.4. Dosimetric Measurements for TPS Validation

a) General Considerations

The dosimetric measurements for TPS validation have two goals; test the accuracy of point dose calculation and interpolation (or analytic computation) procedures. It is good practice to use for the acquisition of control data sets a better spatial resolution than those imposed by the TPS for the acquisition of the basic reference data. The same consideration should be made for the choice of the measuring depths.

Tolerance limits for the calculation of relative dose distribution in region of low dose gradient, are quoted as percentage of the central axis dose in a single field case or as percentage of the dose at the prescription point in a clinical case. In areas of high dose gradient (penumbra, build-up) the tolerance should also be specified in terms of displacement (shift of isodose in cm) of corresponding dose values. The level of acceptance stated in this report is based on the experience of different authors and reflects the precision actually achievable [8],[15],[11],[18],[22]. For each recommended measurement a particular tolerance is given. Deviations in some regions or under some circumstances in excess of the tolerance limits have to be judged in the light of their clinical relevance.

The part concerning the calculation of Monitor Units (MU) is reported in the section concerning the repeated checks. You should also refer to the section 4.2.2 for the commissioning.

b) Photon dosimetry (Photon beam Tests)

The following tests shall be made for each photon beam quality and each therapy unit available.

(i) Beam characterisation data

The aim of this test is to check the basic beam data which is used by the planning system to set up a beam model. Algorithms in commercially available planning systems are based either on tabulated beam data (A) or on analytical treatment unit models (B). Checking beam characterisation data for case (A) means direct checking of the tabulated data by suitable means (graphs, tables, software tools). Errors are not expected and not accepted in these checks. In case (B), beam characterisation data is used to fit some more fundamental model parameters. Some deviations therefore may be expected when beam characterisation situations are recalculated and the calculated values are not necessarily identical to the characterisation data. The following table lists dose differences and isodose shift distances between calculated and measured results, which are considered to be acceptable.

(ii) Open and wedged beams for fields not used for beam characterisation (field sizes of clinical relevance, either interpolated or outside characterisation range)

All these tests are to be performed for the proposed minimal set of fields (assuming that these fields were not used for beam characterisation).

(iii) Irregular, multileaf collimated, asymmetric and dynamically wedged fields

Although all these features may be present in one single field in clinical practice, separated tests should be performed for each of them.

(iv) Co - decay

The output of the reference field for different irradiation times shall be tested.

(v) Inhomogeneity correction, surface irregularities and oblique incidence fields

The dose per monitor unit on the central ray in region of electronic equilibrium shall be tested. Some cases of inhomogeneities are proposed and described in the figure below. Of course, other situations of clinical relevance can also be verified. As the measurements can be difficult [28], it is a good practice to compared with situations found in the literature [24].

figure 1. Description of a proposed testing set

(vi) Standard treatment plans (3D distributions, rotational irradiation, multiple beams)

Several anthropomorphic phantom tests can be used for a final complete test of the entire calculation algorithm. These test cases should be similar to treatment techniques used in the clinic, and of interest to the physicians and physicists involved. Some examples:

c) Electron dosimetry (Electron beam Tests)

For electrons, the dose gradient is steep along the central axis beyond the maximum and at beam edges. In order to avoid uncertainty in the position of the measurement high resolution detectors (diode detector, film, small ionisation chambers) and detectors which have a well defined effective points of measurement should be used (plan parallel ionisation chamber). In the conversion from depth ionisation (ionisation current) to depth dose the SSRMP protocol should be used.

The contents of the test shall include planar dose distributions in a number of planes: (a) a transverse plane containing the central axis of the beam, (b) a transverse plane a few centimetres off axis, (c) planes normal to the beam axis (beam's-eye-view planes) at several depths [33]. Normal plane isodoses verify the fluence distribution over the entire surface which in combination with the depth doses enables beam data verification over the whole volume.

As stated by many authors, the difference between measured and calculated dose could be relatively large (10%) [8],[9],[16] in the case of electron beams. The following tolerances should be regarded as ideal.

The following tests shall be made for each electron beam quality and each therapy unit available.

(i) Beam characterisation data

The aim of this test is to check the basic beam data which is used by the TPS to set up a beam model. Algorithms in commercially available planning systems are based either on tabulated beam data (A) or on analytical treatment unit models (B). Checking beam characterisation data for case (A) means direct checking of the tabulated data by suitable means (graphs, tables, software tools). Errors are not expected and not accepted in these checks. In case (B), beam characterisation data is used to fit some more fundamental model parameters. Some deviations therefore have to be expected when beam characterisation situations are recalculated and the calculated values are not necessarily identical to the input data.

(ii) Open fields not used for beam characterisation (field sizes of clinical relevance, either interpolated or outside the characterisation range)

All these tests shall be performed for a minimal set of field sizes. The field sizes to be included in this set are electron energy and accelerator specific. They could depend also on the calculation algorithm and shall be chosen in order to get an idea where possible weaknesses in the dose calculation would occur. For accelerators using electrons applicators, tests for all of them shall be performed.

(iii) Irregular field shapes, asymmetric and oblique incidence

Situations that would be encountered in the clinical treatment planning should be evaluated: oblique incidence as in the irradiation of the neck or chest wall, irregular geometry as the irradiation of tumours in and behind the nose, fields with low melting point alloy blocking, dose distribution behind bolus.

(iv) Inhomogeneity correction and irregularities of the surface

The dose per monitor unit below slab inhomogeneity of lung and bone density in point of electronic equilibrium shall be tested.

(v) Standard treatment plans

Several anthropomorphic phantom tests can be used for a final complete test of the entire calculation algorithm. These test cases should be similar to treatment techniques used in the clinic, and of interest to the physicians and physicists involved (the nose, mandible, etc.). Particular attention shall be paid to the superposition and matching of field, (also with mixed photon and electron beamsaccording to the TPS capabilities).

Tests after implementation of a new software release

2.1. Introduction

After any software upgrade (installation of new releases or other modification) it is recommended to check the reliability, correctness and constancy of relevant parameters and of calculation algorithm. To be more efficient, a detailed study of the changes expected in the new version, and good documentation of all the expected changes shall be furnished by the vendor [34]. New software releases and upgrades could be classified into two subsets: those not directly influencing the calculation algorithm and those where substantial variation on physical models implemented are introduced (e.g. change in irregular fields treatment, head scatter corrections...)

In this latter case, given the complexity of TPS program structures, the whole calculation algorithm has to be considered changed and as a consequence all involved treatment units shall be checked as already described for commissioning.

In the case of software releases related only to ancillary or to presentation utilities (like graphic tools added or modified), only a subset of the commissioning procedure check are mandatory before the use of the TPS in clinical routine. The following paragraphs are devoted to this case suggesting an explicit verification list.

2.2. Acceptance tests for a new software release not affecting physical algorithms

2.2.1. Non dosimetric checks

All clinically used tools and programs, whether or not involved in the software upgrade, shall be checked: coarse checks, geometry and 3-D functionality, peripherals (see paragraphs 1.2.1 to 1.2.3) need a control as for the commissioning (see paragraph 1.2). It is also important to check for the completeness of the functions, as this may interfere with clinical routine.

2.2.2. Dosimetric checks

a) photon beams

• point dose at 3 depths (d_max, 5 cm, 10 cm) for 5x5 cm², 10x10 cm² and maximum open square fields, a rectangular 5x30 cm² field and an irregular field.

• profiles: for open beams (10x10 cm²  and the maximum square field) at 10 cm depth; for wedged beams only the maximum field allowed at 10 cm depth.

b) electron beams

• point dose for reference field, at 2 depths: d_max and a point in the fall off region

• profiles in the same conditions

c) standard cases

The dose distribution in at least the central beam section and one off-axis section shall be calculated. As recommended in section 1.1.3 (Preliminary check for commissioning), the vendor shall provide a standard and common set of patients and beam data with their output (dose distribution). This set shall be used to test the local implementation of the new software release allowing for inter-comparisons with other TPS. The results obtained from these test patients may be compared with the data obtained from the previous release by means of a systematic dose verification in predefined points and on isodoses distributions.

As an alternative method 3 real patients locally treated could be used to check the new version following the same procedure.

Isodoses, monitor units calculations and DVH results (when available) from the new release, compared with those obtained from the previous one shall match without any quantitative difference unless mentioned as corrections by the vendor.

2.2.3. Verification of all potentialities of the TPS

The standard procedure defined in 1.1.2 shall be tested. This systematic method is intended to reduce necessary time to check all functions, menus and submenus and to avoid doubled checks.

If any task in any TPS related program does not work as expected, it shall be noted in the TPS log-book and precisely reported to the manufacturer customer support teams. The operators shall be informed and temporary alternative procedures set up.

Tests after addition of a new therapy unit and/or implementation of new beam data

The implementation of a new therapy unit or new beam data concerns all steps of a TPS Quality Assurance Program; this implies that in order to exclude any systematic error in treatment planning, all the functions relied to this new treatment unit have to be carefully checked and evaluated before clinical use. From this statement, it follows that a complete set of tests of the software and dose calculation is required as for commissioning. Recommendations described in paragraph 1.2.1 and 1.2.4 apply here.

Repeated TPS checks

4.1. Introduction

The aim is to ensure the constancy of calculation, dose distribution and all outputs from the TPS. Errors can occur from 4 different parts of the whole process of treatment planning : (1)  programs, (2)  beam data, (3)  peripheral devices or (4) operators. Regarding to QA, all these parts shall be periodically tested. The periodic tests shall be made by different operators of the TPS. The idea is that operators can have their own habits in the schedule of executions and consequently use different algorithms in the program and perhaps obtain divergent results. The results could be operator dependent, hence the reason to impose standard procedure for the use of TPS.

Repeated checks as formulated here are an important part of the QAP, but a continued vigilance on the part of the operator is also required. The idea is to recognise the more subtle problems or differences which may occur. Investigation may uncover important issues which shall be resolved [34].

"In vivo" dosimetry is an overall test of the TPS, this subject is left to the judgement of the physicist as it lies outside the scope of these guidelines.

4.2. Tests and frequencies

In this section, the tests are to be performed as described for the commissioning (section 1.2). The frequency of the checks are abbreviated as follow:

4.2.1. Peripherals

You should report to the paragraph mentioned to see more details.

4.2.2. Monitor Units (MU)

A test concerning the calculation of MU shall be performed monthly on clinical cases. See also the reference [6] for more details. The verification of the monitor unit calculation for each patient is not treated in this document, as it is not a part of the TPS QA.

4.2.3. Standard treatment plan

The tolerance given in paragraphs 1.2.4.b.vi and 1.2.4.c.v are relative to the measured dose at the point of normalisation in an anthropomorphic phantom. Here the tolerance (1%) concerns the reliability of the checks done during commissioning. At each point verified the divergence in dose should not exceed 1% of the dose at the point of normalisation.

4.2.4. "Checksum"

The program cannot be modified by the operators therefore it should be verified that it has not been corrupted by any virus or wrong operations. The beam data can be modified (e.g. new measurement more precise). Such modification of the beam data shall be check by the appropriate procedure as describe below. When the beam data are supposed to be definitive, the integrity of the executable files or beam data shall be certified. If a virus corrupts programs or somebody makes a wrong operation on the system, a checksum utility program pointing on the executables, beam data and other important files can identify the problem [4]. It is the task of the vendor to implement it. This recommended test cannot be avoided if periodic tests are performed with standard clinical cases. Checks fulfilled on standard clinical cases only reveal problems without any explanation. On the other hand checksum tests can identify the problem and they can be executed automatically at each start of the TPS (boot procedure). A good practice is to ensure that the checksum is carried out after a registered modification of a parameter. 

[1] AAPM Report No. 55, Radiation Treatment Planning Dosimetry Verification, from the Task Group #23 of Radiation Therapy Committee, Published by the American Institute of Physics, New-York, 1995, ISBN 1-56396-534-8

[2] ACR/NEMA Standards Publication No 300-1988, ACR/NEMA 2.0

[3] British Journal Of Radiology, Supplement 25, Central Axis Depth Dose Data for Use in Radiotheraphy , Published by the British Institute of Radiology, 1996, ISBN 0 905749 38 3

[4] B. Curran, G. Starkschall, A Program For Quality Assurance of Dose Planning Computers, ACMP Symposium "Quality Assurance in Radiotherapy Physics", Galveston, TX, May 1991

[5] Dahlin H. Lamm IL. Landberg T. Levernes S. Ulso N., User requirements on CT-based computed dose planning systems in radiation therapy, Acta Radiologica - Oncology. 22(5):397-415, 1983.

[6] R.A. Dahl, E.C. McCullough, D.E. Mellenberg, A quality assurance program for monitor unit calculators, Med. Phys. 17 (1), 1990, 103-105

[7] ACR/NEMA Standards Publication PS3, DICOM3

[8] J. Van Dyk, R.B. Barnett, J.E. Cygler, P.C. Shragge, Commissioning and Quality assurance of treatment planning computers, Int. J. Rad. Onc. Biol. Phys. 26, 1993, 261-273

[9] K. Friedrich, M. Fippel, « Externe Dosisberechnung mit TMS-Planungsdaten, Z. Med. Phys. 6, 163-168, 1996

[10] ICRU Report 35, Radiation Dosimetry: Electron Beams with Energies Between 1 and 50 MeV, International Commission on Radiation Units and Measurements, Maryland, USA, 1984

[11] ICRU Report 42, Use of Computers in External Beam Radiotherapy Procedures with High-Energy Photons and Electrons, International Commission on Radiation Units and Measurements, Maryland, USA, 1987

[12] ICRU Report 50, Prescribing, Recording, and Reporting Photon Beam Therapy, International Commission on Radiation Units and Measurements, Maryland, USA, 1993

[13] IEC 62083, "Equipment for radiotherapy, nuclear medicine and radiation dosimetry", specially Ed. 1: "Electromedical equipment - Particular requirements for the safety of radiotherapy treatment planning systems", by the International Electrotechnical Commission, 30.10.1998 version

[14] J. Jacky, C. P. White, Testing a 3D radiation therapy program, Int. J. Rad. Onc. Biol. Phys. 18, 1990, 253-261

[15] T. Knöös, C. Ceberg, L. Weber, P. Nilsson, The dosimetric verification of a pencil beam based treatment planning system, Phys. Med. Biol. 39, 1994, 1609-1628

[16] A. Kosunen, H. Järvinen, S. Vatnitskij et al., Intercomparison of radiotherapy treatment planning systems for external photon and electron beam dose calculations, Radiotherapy and Oncology 29, 327-335, 1993

[17] G.J. Kutcher, L. Coia, M. Gillin, W.F. Hanson, Comprehensive QA for radiation oncology : Report of AAPM Radiation Therapy Committee Task Group 40, Med. Phys. 21 (4), 1994, 581-618

[18] McCullough EC. Krueger AM., Performance evaluation of computerized treatment planning systems for radiotherapy: external photon beams, Int.J.Rad. Oncol. Biol. Phys 6(11), 1980, 1599-1605

[19] P.C. Shrimpton, D.G. Jones, M.C. Hillier, B.F. Wall, J.C. Le Heron, K. Faulkner, NRPB Report R248, Survey of CT Practice in the UK, Part 1, by the National Radiological Protection Board, Oxon, UK, 1991

[20] P.C. Shrimpton, D.G. Jones, M.C. Hillier, B.F. Wall, J.C. Le Heron, K. Faulkner, NRPB Report R249, Survey of CT Practice in the UK, Part 2, by the National Radiological Protection Board, Oxon, UK, 1991

[21] Three-dimensional photon treatment planning report of the collaborative working group on the evaluation of treatment planning for external photon beam radiotherapy, Int. J. Rad. Onc. Biol. Phys. 21 1, 1991

[22] IPEMB Report 68, «A guide to commissioning and quality control of treatment planning systems», 1996, by the Institution of Physics and Engineering in Medicine and Biology, York, England, ISBN 0904 181 839

[23] U. Rosenow, Qualitätskontrolle in der Bestrahlungsplannung, Z. Med. Phys 1 (1991) 59-67

[24] C. Kappas, J.C. Rosenwald, Quality control of inhomogeneity correction algorithms used in treatment planning systems, Int. J. Rad. Onc. Biol. Phys. 32 3 (1995), 847-58

[25] Evaluation des performances et contrôle de qualité des scanneurs", juin 1990

[26] D.L. McShan, Comments on: Commissioning and Quality assurance of treatment planning computers, Int. J. Rad. Onc. Biol. Phys. 26, 1993, 371-372

[27] Recommandations No 1 : «Contrôles physiques et dosimétriques en téléradiothérapie», Société Suisse de Radiobiologie et de Physique Médicale (1982, rév. 1992)

[28] R.L. Stern, B. A. Fraass, A. Gerhardsson, D. L. McShan, K.L. Lam, Generation and use of measurement-based 3-D dose distributions for 3-D dose calculation verification, Med. Phys. 19 (1), 1992,165-173

[29] G.K. Svensson, Quality assurance in radiation therapy: physics efforts, Int. J. Rad. Onc. Biol. Phys. V10, Sup 1, 1984, 23-29

[30] H. Svensson, Quality assurance in radiation therapy: physical aspects, Int. J. Rad. Onc. Biol. Phys. V10, Sup 1, 1984, 59-65

[31] C.Westermann CF. Mijnheer BJ. van Kleffens HJ,Determination of the accuracy of different computer planning systems for treatment with external photon beams, Radiotherapy & Oncology. 1(4):339-47, 1984 Mar

[32] D.R. White and R.D. Speller - "The measurement of effective photon energy and linearity in computerized tomography" Brit. J. Radiol. 53, 1980, 5-11

[33] Shiu AS, Tung S, Hogstrom KR, Verification data for electron beam dose algorithms, Med. Phys.19,623-636 (1992)

[34] Teletherapy: Present and Future, Proceedings of the 1996 summer school, «Quality assurance for 3D treatment planning», p. 253-318, AAPM, ISBN: 1-888340-03-7

The two following tables represent a summary of all the tests that should be performed for the commissioning, respectively repeated QA procedures.

Commissioning checks (see chapter 1)

(1) In case of disagreement, a specific procedure should be implemented
(2) The check apply for any entered data
(3) The check shall attest that the calculations based on the entered data are correct (see also section 1.2.1)
(4) The case of electronic densities is given as an example (see related sections)
(5) Minimal set of field sizes : 5x5,10x10,25x25,40x40 (max. square field), 5x10, 5x25,5x40, 10x5, 25x5, 40x5

Repeated checks (see chapter 4)

(1) This check shall also be executed at each revision of the CT.
(2) This check consist in a verification of the data transfer.
(3) This check should be included in the boot procedure of the TPS.

Footnotes:

1. The operator designates the person which performs the routine planning on the TPS

2. Treatment unit model: All physical and radiation parameters, for a particular piece of medical electrical equipment, that are needed to plan a course of radiotherapy, including dose calculation [13]

3. Manufacturer could also be mentioned here, but vendor is preferred considering that he is the responsible for the correct functioning of the TPS.

4. As example, a treatment unit model without the wedge data, or without data concerning some energies available on the actual treatment unit.

5. The term user shall be understand in the meaning of institution (e.g. the hospital)

6. Dose per Monitor Unit is a more general term than output factor (see next note).

7. The British Journal of Radiology [3] defines the output factor as the absorbed dose rate at the point on the beam axis at the depth d_m for a given fiel size, normalised to unity for a specified standard field size. Relative output factor can be considered as a synonym for output factor.

SGSMP Home Page

This page is maintained by Pierre-Alain Tercier / Last updated: 22 September 2011(wws).
Please forward any comments and/or additions to this webpage to: Webmaster SGSMP.