School on Medical Physics for Radiation Therapy:

Dosimetry and Treatment Planning for Basic and Advanced Applications

Trieste - Italy, 25 March-5 April 2019

Treatment Time / MU
calculation in RT
Maria Rosa Malisan
 Clinical Dose Calculations
 Computing absorbed doses in a patient using data measured
in a phantom has been the standard of practice in
radiotherapy (RT).
 This is because direct measurement of absorbed doses in a
patient is impractical and often impossible.
 Therefore, the treatment planning has to be based on
calculation models.
 Even if direct measurements were possible, it would still be
much more practical and convenient to perform planning
based on calculation models.
 The dose predicted by a calculation method should
correspond to the real absorbed dose in the patient as
accurately as possible. 2
 RT Planning
 In RT treatment planning, the purpose is to devise a
treatment, which produces as uniform dose distribution as
possible to the target volume and minimizes the dose
outside this volume.

3
 RT Planning
 In RT planning, the beam qualities, field sizes, positions,
orientations and relative weights between the fields are typically
modified.
 It is also possible to add certain accessories (e.g. wedge filters or
blocks) to the fields to account for oblique patient surface or to
shield critical structures from radiation exposure.

4
 Historical Background
• Practising of treatment planning started in
1940’s when the developments in radiation
dosimetry enabled each clinic to measure
the isodose charts for any type of treatment
field, thus enabling manual 2D planning.
• To avoid laborious isodose measurements,
empirical methods for the calculation of
dose distribution were developed later.
• e.g. the percent depth dose (PDD) was
introduced to calculate doses for treatments
delivered using fixed treatment distance
machines.
5
 Historical Background
• Computer-based treatment planning systems (TPSs), first
introduced in the ‘70’s of last century, allowed the planner to see
the effect of the beam modifications immediately on the predicted
dose distribution.
• This resulted in better quality plans, since it became easier to
experiment with a larger set of treatment parameters.
• Moreover, it improved dose-calculation accuracy with the
incorporation of patient-specific anatomical information.

6
Factor vs Model-based algorithms
• First TPS’s made use of factor based
models, where the dose per MU is
typically expressed as the dose to a
reference point under reference
conditions, corrected with a set of
factors.
• Each factor accounts for one or several
different effects:

– beam size, beam shape, depth, distance, wedges, etc.

• These factors are typically measured or calculated through simple
modelling and stored in tables.
• The method is intuitive and robust, but lacks general applicability.
• It is in principle impossible to account for all different treatment
design possibilities which are a part of modern radiotherapy. 7
 Modern Treatment Planning Systems
• Therefore the model-based calculation methods were introduced
within TPS’s, where the commissioning measurements are used to
determine a set of more fundamental physical parameters which
characterize the radiation from the treatment unit.
• Model based algorithms can be made fully general without the
need for a large set of characterization measurements.
• Recently, 3D TPS’s have become common in RT departments
offering improved accuracy and enhanced visualization in the RT
treatment planning process.
• With recent improvement in computing technology, the newer TPS
now correctly model the radiation transport properties three
dimensionally and estimate the dose deposition precisely.
8
 MU calculation
• In external beam RT, monitor units (MU) or beam-on time for a
given treatment plan allows the RT technologists to deliver the
actual dose to a patient.
• MU are calculated by the TPS by means of sophisticated algorithms
from the calculated dose distribution and dose prescription.
• It is essential for the user of a TPS to understand the principles of
the MU calculation algorithm!
• However, in “simple” cases MU can be computed by means of
several dosimetric functions introduced to relate absorbed doses
measured in a phantom to absorbed doses in a patient:

Manual calculation

9
 Why Manual MU Calculation ?
 Traditionally manual calculation is carried out by means of
(correction) factor- based models.
 It can sound utterly out of fashion in the era of physics-based
models or Monte Carlo TPS !
 However, it can result useful as a powerful QA tool during TPS
commissioning.
 In fact, modern model-based TPS’s dose calculations, make use of
characterization measurements to determine more basic
parameters: errors in characterization measurements can result in
unexpected and systematic calculation errors.
 Moreover, software errors can go undetected during commissioning
and manifest subsequently in clinical planning

10
 Why Manual MU Calculation ?
• ICRP Report 86 has categorised 46 accidents/incidents reported for external
radiotherapy as categorized by ICRP 86
accidents reported in ext RT:
28% in treatment planning and dose
calculation.
• The human factor is the cause for a
large majority of the incidents and
accidents. In routine clinical practice,
more likely sources of systematic
dose error for individual patients
result from a lack of:
– understanding of the TPS;
– appropriate commissioning (no
comprehensive tests);
– independent calculation checks.
T. Nyholm,
112008
List of reported bugs from the TPS vendors collected from the
FDA MAUDE database for the time period 2004‐2008.

• The companies are not

obligated to report all
problems, and different
companies have
different policies
regarding the reporting.

• The presented list of

identified bugs are
therefore far from
complete and is
perhaps not even
representative.
T. Nyholm, 2008
12
 Why Manual MU Calculation ?
• The major issues that relate to treatment planning errors can
be summarized by four key words:
Education; Verification; Documentation; Communication

• The ICRP Report 86 concluded that many of these accidents

could have been prevented through independent
verification of the TPS and with systematic use of in‐vivo
dosimetry.
• Independent verification can also enhance confidence in the
accuracy of the algorithm and integrity of the beam data
used.
• It may also be a formidable didactic tool!
13
 MU Calculation for TPS Commissioning
• IAEA TRS 430 Report lists some
of the relevant issues that
should be investigated
• It briefly describes the types of
test that can help to verify the
correct behaviour of the entire
planning and MU/time
calculation process.
• Detailed checks of the entire
planning and MU/time
calculation process should be
performed.
14
IAEA TRS 430:
MU calculation tests
• A number of important
aspects of the treatment
planning process affect the
way one should calculate
the MU’s or time (e.g.
normalization)

• For these 9 test situations,

the MU/time calculation
performed using the TPS
should be compared to the
manual MU/time
calculation.

15
 IAEA TRS 430: Overall Clinical Tests
• Measurement or
manual dose
evaluation of the final
dose delivery should
be performed,

• to ensure that the

correct absolute dose
would be delivered to
the patient following
the completion of the
total treatment
planning process.

• While it is not necessary to implement these particular examples, it is

important that some typical situations be developed and tested right
through to the evaluation of absolute dose. This is especially true for a
new TPS. 16
• MU’s calculated using the TPS were compared with MU’s calculated from
point dose calculations from TMR tables.
• Discrepancies in MU calculations were both significant (up to 5%) and
systematic:
• 1) a coordinate system transformation error,
• 2) mishandling of dose-spread arrays,
• 3) differences between dose calculations in the commissioning software and
the planning software,
• 4) shortcomings in modeling of head scatter.
• Corrections were made in the beam calculation software or in the data sets
to overcome these discrepancies. Consequently, we recommend validation
of MU calculations as part of commissioning process. 17
 Independent Dose calculation
• Dose calculation with a TPS represents one of the most
critical links in the RT treatment process, since it is the only
realistic technique to estimate dose delivery in situ.
• Even though the calculation algorithms are tested during the
commissioning of TPS and results are achieved with 1-2%
accuracy in water phantom geometry, a good QA programme
further requires that

all MU’s calculated for clinical use should be

verified using a second independent calculation method

• so that any errors due to software faults and improper use of

the systems could be identified.
• This check becomes more important as the sophistication of
the planning algorithm increases.
18
 Safety Legislation
• In several European countries there are legal aspects based on
EURATOM directive 97/43 for independent QA procedures and
their implementation into national radiation protection and patient
safety legislation.
• In particular, Article 8 states: “Member States shall ensure that…
appropriate QA programmes including quality control measures and
patient dose assessments are implemented….”.
• This is also emphasized in Article 9 with respect to Special Practices:
“…special attention shall be given to the QA programmes, including
quality control measures and patient dose or administered activity
assessment, as mentioned in Article 8.”
• In a broad sense this directive directs the holder to assure that the
delivered dose to the patient corresponds to the prescribed dose. 19
 Independent MU calculation
 Dose errors arising in computing the MU could potentially affect the
whole course of treatment and therefore are of particular concern.
 So, independent checking of MU calculations, for each RT
treatment plan, is essential for QA.
 It is considered more than desirable if the beam data set and
calculation algorithm are independent of those of the TPS.
 AAPM also recommends an independent calculation of the dose at
one point in the plan, preferably at the isocenter or at a point near
the center of the PTV.
 If the independent calculation differs from the treatment plan by
more than a pre-set tolerance level, the disparity should be
resolved before commencing or continuing treatment.
20
 Independent dose calculation
 Dose calculations can be performed through various methods
using fairly different approaches.
 A tool for independent dose calculations is a compromise
between the benefits and drawbacks associated with different
calculation methods in relation to the demands on
accuracy, speed, ease of use.
 Independent dose calculations have been used for a long time
as a routine QA tool in conventional RT using empirical
algorithms in a manual calculation procedure, or using
software based on fairly simple dose calculation algorithms
 (Dutreix et al., 1997; Knöös et al., 2001; van Gasteren et al., 1998).
21
 MU Verification: ESTRO and AAPM docs
 Recommendations for MU verification have been published by
ESTRO (Booklets 3 and 6) and by the Netherlands Commission on
Radiation Dosimetry, NCS .
 AAPM Task Group 71, formed in 2001 to create a consistent
nomenclature and formalism (national protocol) for MU
Calculations, published the Report 258 In 2014 :
 Monitor unit calculations for external photon and electron beams:
Report of the AAPM Therapy Physics Committee TG No. 71, Medical
Physics, Vol 41, Issue 3
 In these reports it is common practice to verify the dose at a point
by translating the treatment beam geometry onto a flat
homogeneous semi-infinite water phantom or “slab geometry”.
 Users should be aware of the limitations of this compromise that
favors simplicity and calculation speed over accuracy!
22
 Limits of the manual MU verification
 Conventionally, MU
calculation verification
methods assume ‘‘water
phantom geometry’’ in which
the beam is presumed to be
incident on a slab of material
affording full scatter
conditions.

 It is evident that this

assumption yields over- or
under- estimated scatter
contributions, depending on
the exact geometry.

ESTRO Booklet 10 23
 Limits of the manual MU verification
 Various methods to handle and
correct for density variations
(heterogeneities) in the literature
 Most often these heterogeneity
corrections rely on one-
dimensional depth scaling along
ray lines from the direct source,
employing equivalent/ effective/
radiological depths that replace
the geometrical depths.
 In general, the full 3D nature of the
process can not be properly
modelled.
 The result is that all deviations
from the ideal slab phantom
geometry will cause different
errors in the calculated doses. ESTRO Booklet 10 24
 Manual MU Verification experiences

 An independent MU calculation is created in an MS-Excel

spreadsheet. The method is shown sufficiently sensitive to identify
significant errors and is consistent on the magnitude of
uncertainties in clinical dosimetry.

 It is reported that using straightforward but detailed computer

based verification calculations, it is possible to achieve a precision
of 1% when compared with a 3D Helax TPS MU calculation.
25
 the MU’s calculated by Pinnacle planning system were compared
with hand calculations from lookup tables for nearly 13,500
treatment fields without considering the tissue inhomogeneity.
 The 3D TPS MU calculation was systematically higher than the
‘‘hand’’ calculation: for simple geometries the mean difference was
1% and was as high as 3% for more complicated geometries.
 Careful attention to factors such as patient contour could reduce
the mean difference.
 ‘‘Hand’’ calculations were shown to be an accurate and useful tool
for verification of TPS MU calculations. 26
 Manual Verification experience

 This system has been implemented into the daily clinical quality control
program.
 A hand-held PC allows direct calculation of the dose to the prescription
point when the first treatment is delivered to the patient.
 The model is validated with measurements and is shown to be within ±1.0%
(1 SD).
 Comparison against a state-of-the-art TPS shows an average difference of
0.3% with a standard deviation of ± 2.1%.
 An action level covering 95% of the cases has been chosen, i.e. ± 4.0%.
 Deviations larger than this are with a high probability due to erroneous
handling of the patient set-up data.
27
28
 Factor-based dose calculation
 Traditionally the most common way of calculating the dose is
through a series of multiplicative correction factors that describe
one-by-one the change in dose associated with a change of an
individual treatment parameter, such as field size and depth,
starting from the dose under reference conditions.
 This approach is commonly referred to as factor-based calculation
and has been the subject of detailed descriptions.

29
 Factor-based dose calculation
 The individual factors are normally
structured in tables derived from
measurements or described
through parametrizations.

 Some factors can be calculated

through simple modelling, for
example the inverse square law
accounting for varying treatment
distances.

 From an implementation point of

view a factor-based method may
be an attractive approach due to
its computational simplicity, once
all the required data are available.
ESTRO Booklet 6 30
 Limits of the Factor-based dose calculation
 The obvious problem associated with this approach is the
required amount of commissioned beam data as this type of
method can not calculate doses when the beam setup is not
covered by the commissioned set of data.

 For treatment techniques that can make use of many degrees

of freedom, such as the shape of an irregular field, it becomes
practically impossible to tabulate or parameterize all factors
needed to cover all possible cases.

 Hence, the factor-based approach is best suited for point dose

calculations along the central beam axis in beams of simple
shapes and simple modifiers (wedges, blocks, MLC…).
31
Independent MU
calculation: suggested
steps by NCS (2005)
a. Develop a MU calculation program, either for manual calculation or using a
computer program, based on the formalisms given in ESTRO Booklets 3 and 6 or
NCS Report 12. See also Venselaar et al.
b. Include in the program the dependence on depth (using the percentage depth-
dose, PDD, or tissue-phantom ratio, TPR), SSD, field size, and preferably taking
the collimator exchange effect into account.
c. Take into account the dose variation with field size in case of the presence in the
beam of a wedge or a blocking tray by using field size dependent correction
factors.
d. For more complex situations involving tissue inhomogeneities, off-axis situations
and MLC-shaped fields, more sophisticated algorithms are required. Several
groups are currently in the process of developing these algorithms. 32
https://www.estro.org/binaries/content/assets/estro/school/publications/booklet-
10---independent-dose-calculations---concepts-and-models.pdf

33
34
 MU Verification Software
 The manual calculations are expected to be less accurate than
those performed by the TPS because factors such as patient
surface convexity, tissue heterogeneity or beam obliquity are not
considered.

 Moreover, with the introduction of Intensity Modulation

Radiation Therapy (IMRT), an independent manual calculation of
MU becomes difficult due to the complex relationship between
the MU and the beam shape as well as the technique used to
generate the intensity modulation.

 Currently, a variety of new MU verification

software packages have been introduced
in the market and are claimed to be
capable of accurately calculating the MU’s
even for IMRT.
35
 IMSURE QA

RADCALC

MUCHECK
Conclusion: the variation of the MU calculations
between the examined software was found to be very
similar indicating that their ability to be used as QA tools
of the TPS calculations is equivalent. 36
DIAMOND
 Evaluation and comparison of second-check monitor unit
calculation software with Pinnacle3 treatment planning system
B. Tuazon, G. Narayanasamy, N. Papanikolaou, N. Kirby, P. Mavroidis, S. Stathakis

Abstract
• The purpose of this study was to evaluate and compare the accuracy of dose
calculations in 2nd check softwares (Diamond, IMSure, MuCheck, and
RadCalc) against the Phillips Pinnacle3 TPS.
• …..
• The mean percent difference in calculated dose for Diamond, IMSure,
MuCheck, and RadCalc from Pinnacle3 were −0.67%, 0.31%, 1.51% and
−0.36%, respectively.
• The corresponding variances were calculated to be 0.07%, 0.13%, 0.08%, and
0.03%; and the largest percent differences were −7.9%, 9.70%, 9.39%, and
5.45%.
• The dose differences of each of the second check software in this study can
37
vary considerably and VMAT plans have larger differences than IMRT. [….]
AAPM TG114:
Computer-based MU verification programs
 Most computer-based MU verification programs use an automated
table look-up method similar to that outlined for manual
calculation, e.g. in ImSure software:

RxDose / IsoDoseLine
MU=
TMRxOCRxWFxTFxSc( FS ) xSP( FS ' ) xCFxUFxInvSqCorr

 Some more complex MU calculation programs use pencil beam or

convolution/superposition algorithms based on the empirical data.

 These computer programs require commissioning at multiple

points and periodic QA to verify the continued data integrity and
calculation algorithm functionality.
38
 AAPM TG114: Guidance for Action Levels

 Recommendations on establishing Action Levels for

agreement between primary calculations and verification,
and guidance in addressing discrepancies outside the action
levels are provided.

 These recommendations shall not be interpreted as

requirements.

 It is important that the physicist knows the accuracy and

limitations of both the primary and the verification systems
in order to set reasonable and achievable action levels and
to better interpret the causes of differences between the
two results.
39
 AAPM TG114: Guidance for Action Levels
 The level of agreement achievable depends on the details of
the patient geometry, the primary and the verification
calculation programs, and the clinical situation, in addition to
whether corrections for tissue heterogeneities are used.

 It is therefore reasonable to have different action levels for

different situations.

 Each institution must determine the proper action levels for

that particular clinic.

 Results from planning system commissioning are useful in

establishing these levels.
40
 AAPM TG114: Guidance for Action Levels

 A base action level of 2% was postulated for simple field

geometries, consistent with the AAPM TG-53 criterion of 2%
dose accuracy between calculations and measurements.

 From this starting point, additional range was added to

account for the increased uncertainties of complex treatment
geometries.

 The action level guidelines are divided into two tables,

depending on whether or not tissue heterogeneities are taken
into account in the primary calculation.

41
AAPM TG114: Guidance for Action Levels

42
AAPM TG114: Guidance for Action Levels

 When a discrepancy is noted, the first action should be to

verify that a calculation error has not been made.

 If this basic review fails to identify the cause of a

discrepancy, the next step should be to confirm that an
appropriate comparison point has been chosen.

 Differences in accounting for patient geometry between the

primary and the verification calculations can also lead to
large discrepancies between results (e.g. breast treatment).

43
 AAPM TG114: Guidance for Action Levels

 Density corrections are required for verification of

calculations which include heterogeneity effects.

 The verification calculation must at least take into account

the radiological thickness of tissues overlying the point of
calculation.

 At a minimum, if a discrepancy is attributed to differences

in the calculation algorithms, an assessment to confirm that
the discrepancy is the correct order of magnitude and
direction should be made.

44
 Conclusions-1

 «Manual» MU/time calculation can still have a role in

modern RT characterised by sophisticated computation
algorithms and 3D complex patient models.

 A measurement-based algorithm can have a good didactic

value since it enables to decompose a calculation and
consider the impact of each factor on an individual basis.

 It can be of value during commissioning of clinical model-

based TPS’s , as required by the IAEA TRS 430.

45
 Conclusions-2
 It results an essential tool in the “independent second
check” for MU’s or time calculated to deliver the
prescribed dose to a patient, where a key aspect is the
independent nature of the calculation methodology and
of the beam data and treatment parameters.

 However, its effectiveness in clinical practice relies on a

proper commissioning in order to assess its accuracy and
limitations, so to set reasonable action levels and to
better interpret the causes of differences between the
two calculations.
46
Manual Calculation Tools

47