Practise guideline 5a Flashcards
how long should TPS commissioning take?
with 1.5-2 full time physicists working on it,
2-4 weeks for single energy photon beam
4-6 weeks for 2 photon energies plus 5 electron energies
what does TPS commissioning include?
beam data acqisition, modeling, and verification
flow of steps in TPS commissioning
acquire data- review data- process data- review data- then iterative model
-iterative model includes basic photon validation, heterogeneous photon validation, IMRT/VMAT validation, electron validation
-then goes to documentation
-then to routine QA
is stereo included in scope of this practise?
no
small fields
measure down to 2x2 cm2 with appropriate detector
MLC measurements
intra/interleaf leakage with film
leaf end penumbra with a small chamber to avoid volume averaging
when do you review data
before and after entry into planning system
what data review is done prior to entry into TPS?
Inverse square effects, beam divergence, expected beam energy changes with field size, and other well-known characteristics should be validated (this is often easily performed by review of graphical display of the results). Crossbeam profiles at varying depths and field sizes can be superimposed on the same plot to identify trends.
Depth-dose plots can be analyzed in a similar fashion.
-compare data to reference or nearly identical machine
-validate middle and extreme values
what is involved in data review after data is entered into TPS?
potential processing errors (e.g., problems during import, smoothing, mirroring). A combination of graphical review and spot-checking can be used.
how is beam modelling performed?
according to vendor instructions
even if one is provided, have to validate it and check with reference
-iterative process
TPS model comparison tests and tolerances for basic photons
-dose distribution in a large field should be identical (to within statistical uncertainty) for planning and physics model
-dose in test plan vs calibration condition should be within 0.5%
-Dose distribution calculated in planning
system vs commissioning data (i.e. PDD and OARs for various field sizes and deoths) within 2 %
basic photon validation tests
-small MLC shaped field (non SRS)
-large MLC shaped field with blocking (ex mantle)
-off-axis MLC shaped field, with max allowed MLC over-travel
-asymmetric field at minimal anticipated SSD
-10x10cm2 field at oblique incidence
-large (>15 cm) field for each non-physical wedge angle
basic photon evaluation test tolerance in high dose region
2% if one parameter changed from reference
5% if more than one parameter changed from reference
basic photon evaluation test tolerance in penumbra region
3 mm DTA
basic photon evaluation test tolerance in low dose tail region
3% of max field dose up to 5 cm from field edge
how to do basic photon tests for an inverse-only planning system like tomotherapy?
For an inverse planning only TPS (e.g., tomotherapy), the basic photon tests can be performed by creating simple targets and optimizing a plan for each case (e.g., small, large, on/off-axis, variable SSD) or by using computed data for static fields provided by the vendor.
why do you need to recheck basic photon evaluation after doing IMRT/VMAT modelling?
additional modeling for the IMRT/VMAT may affect the parameter results of the basic photon beam modeling, specifically the penumbra and tails.
3 regions relevant for heteorogeneity corrections that need particular care
1)within low density tisuse
2) near the interface of heterogeneous tissues
3) beyond low/high density tissue
what does commissioning of heterogeneity corrections require?
accurate commissioning of the beam itself and accurate characterization of the patient data
dose to medium or dose to water?
Monte Carlo and GBBS algorithms directly calculate dose to the material within the voxel (“dose to medium”). This can be converted to “dose to water” through application of stopping power ratios, with the goal of reproducing conventional (e.g., C/S) TPS doses.(33) However, this stopping power-based conversion has actually been found to decrease dosimetric agreement with conventional TPS doses in most cases,(34,35) leading to “dose to medium” being recommended.(34)Nevertheless, further study is warranted and vendors are currently encouraged to calculate both dose to medium and dose to water; the QMP must be aware of which dose is being reported
validation of heterogeneity commissioning
-confirm CT density table
-heterogeneity corrections before and after heterogeneity, outside of build-up region- tolerance is 3% for lung
-any heterogeneous phantom can be used
details regarding how to do heterogeneity tests for dose values above and below heterogeneity
i.Measurements should be made outside of the buildup/builddown regions.(26) This simple test allows for the direct study of the calculation accuracy through the heterogeneity.
ii.The recommended field size is 5 × 5 cm2 because discrepancies due to low-density material tend to be exacerbated at smaller field sizes.
iii.Further tests deemed appropriate by the QMP to challenge the accuracy of the particular calculation algorithm being employed should be used to bring a better understanding of the limitations of dose calculation in the vicinity of heterogeneities.
recommended algorithms for heterogeneous media
-pencil beam not acceptable
-collapsed cone
-convolution superposition
-monte carlo
-boltzmann
why do we acquire small field PDD even if TPS doesn’t use it?
TPS will extrapolate so want to check what its limitations are
VMAT/IMRT tests
-verify small field PDD (< 2x2 cm2)
-verify output for small MLC defined fields (not a clinical MLC defined field by itself- testing output of an MLC closed beyond jaw)
-plan, measure, and compare results to TG119 report for head and neck and C shaped cases
-plan, measure, and analyze results for at least 2 relevant clinical cases
-simulate, plan, and treat, with anthropomorphic phantom with embedded dosimeters- send dosimeters out for external review
-if the facility is planning to employ IMRT/VMAT in the thoracic region, a second end-to-end test with a heterogeneous thoracic phantom should be performed.(39) Even with modern model-based dose calculation algorithms, systematic differences between calculated and measured doses in lung have been noted(40,39)and can be worsened by user-configurable parameters.
-If mailing is not available, at least do review with independent medical physicist
VMAT/IMRT evaluation methods and tolerances
-ion chamber measurement in low gradient target region- 2 %
-ion chamber measurement in OAR region- 3 %
-planar/volumetric array- 2%/2 mma, no pass rate tolerance,but areas that do not pass need to be investigated
-E2E- 5% of target dose in low gradient region
why 2%/2mm for array?
3%/3mm may mask some issues that can be easily fixed with commissioning
algorithms used in electron TPS
pencil beam or monte carlo
electron TPS validation tests with tolerances
-Plot PDD and output factors for all cones (with standard cutout sizes) for each energy to confirm the correct qualitative behavior as a function of field size and energy
-isodose distribution- custom cutouts at standard and extended SSDs: 3%/3mm
-suface irregularities and oblique incidence -use reference cone and nominal clinical SSD- 5 %
-inhomogeneity tests- reference cone and nominal SSD- 7%
-Clinically used nonroutine electron setups (e.g., abutting electron/electron fields, electron/photon fields, and small fields that results in a loss of lateral electron equilibrium) will require additional dosimetric verification
goals of TPS QA program long-term
1) the TPS has not been unintentionally modified, and 2) dose calculation is consis-tent following TPS upgrades
how to identify unintentional modifications?
file integrity checksums
how to check TPS consistency
-repeat subset of commissioning tests annually or after TPS upgrades
typical annual TPS QA
-Re-calculate reference plans
-include extended SSD and bolus in electron plans
-can test thorax data set
-All routine QA recalculations should agree with the reference dose calculation to within 1%/1 mm. A partial or complete recommissioning (including validation) may be required if more significant deviations are observed
