Measurement Issues in Commissioning and
Benchmarking of Monte Carlo Treatment
Planning Systems
Jan Seuntjens
McGill University
Montreal, Canada, H3G 1A4
jseuntjens@medphys.mcgill.ca
Outline
• Dosimetric accuracy requirements in
treatment planning
• Measurement dosimetry fundamentals
• Measurement issues in commissioning
• Measurement issues in verification
• Conclusions
AAPM Summer School, 2006
1
Clinical QA
• Clinical QA protocols & guidelines
– AAPM TG 53 (Fraass et al 1998)
– IEC 62083 (International Electrotechnical
Comission, 2000)
– ESTRO (Mijnheer et al 2004)
– NCS (2005)
– etc.
• Content of clinical QA
– dosimetric verification and consistency
– many other aspects of system QA and consistency
AAPM Summer School, 2006
2
Dosimetric accuracy
• effect of dose
uncertainty on
complication-free
tumour control (“utility
function”, Schultheiss &
Boyer, 1988)
• ICRU Report 24 (1976)
• Mijnheer et al (1987)
• Brahme (1988)
1% higher accuracy -->
2% increase of cure
5%
3.5%
3%
The structures these numbers are referring to are targets.
What about OAR’s?
AAPM Summer School, 2006
3
Dosimetric accuracy (cont’d)
Present (1
2.0
Possible (1
)?
1.0
1.1
0.5
Monitor stability
1.0
0.5
Beam flatness / symmetry
1.5
0.8
Patient data uncertainties
1.5
1.0
Beam and patient setup
2.5
1.6
Overall (excluding TPA)
4.1
2.4
Dose calculation (TPA)
1.0 / 3.0 / 5.0
0.5 / 2.0 / 4.0
Overall (including TPA)
4.2 / 5.1 / 6.5
2.4 / 3.1 / 4.7
Absorbed dose at reference point in
clinic
Absorbed dose at other points
)
TPA: treatment planning algo.
AAPM Summer School, 2006
4
Dosimetric and positional accuracy of TPA:
acceptability criteria
Homogeneous
water slab simple fields
Stack of tissue
slabs simple fields
antropomorphic
phantom and/or
complex beams
central
axis
high dose
region low
dose
gradient
large dose
gradient
2%
3%
4 mm
3% (50%
loc.dose)
3%
3%
4 mm
3% (50%
loc.dose)
4%
4 mm
3% (50%
loc.dose)
low dose
region
low dose
gradient
Van Dyk (1993)
AAPM Summer School, 2006
5
Gamma analysis for testing the acceptability
of dose distributions by comparing
measurements and calculations
Illustration of the gamma concept (Low et al 1998) for comparison of MC calculated (a) and film
measured (b) two dimensional dose maps. (c) illustrates the gamma-map for tolerance criteria of
3% (dose), 5 mm (distance).
⎛ d (i ) ⎞ 2 ⎛ r (i ) ⎞ 2
⎟⎟ + ⎜⎜
⎟⎟
γ (i ) = min ⎜⎜
⎝ Δd ⎠ ⎝ Δr ⎠
d (i ) = Dm (i ) − Dc
r (i ) = rm (i ) − rc
AAPM Summer School, 2006
for all points (Dc,rc)
6
Clinical measurement dosimetry
Classification of dosimeters
Dosimeter
Mechanism
Gas-filled ionization chamber Ionization in gasses
Liquid ionization chamber
Ionization in liquids
Semiconductors (diodes,
diamond detectors)
TLD
Ionization in solids
Scintillation counters
Fluorescence
Film, gel, Fricke
Chemical reactions
Calorimetry
Heat
Luminescence
AAPM Summer School, 2006
8
Measurement dosimetry in water
where:
dose to water at the point
raw signal, corrected for environmental
conditions as given by detector
detector cavity dose calibration coefficient
(coupling constant)
dose conversion coefficient converts average
detector dose into dose to water at point
AAPM Summer School, 2006
9
Dosimeter dependent coefficients
& coupling constants
Factor or
Dosimetry technique
coefficient Calorimetry Fricke
Ionchamber
dosimetry dosimetry
R
cdet
fmed
ΔT
C
Unity
ΔOD
ρL
1
( Fe3 + )G
ε
med
D
( )Fricke
AAPM Summer School, 2006
Q
1 Wgas
mgas e
smed,gasp Q
10
Interpretation of a dosimeter
measurement
Dmed = Ncav Ms med ,cav pQ
Dcav
fmed
• Ncav is the cavity dose calibration factor
and ties the clinical dose measurement
to the chamber/detector calibration
• M is the result of a measurement corrected
for “technical” influence quantities to ensure
that what we actually measure a quantity
proportional to the dose to the detector
material
AAPM Summer School, 2006
11
Interpretation of a dosimeter
measurement (cont’d)
Dmed (r) = N cav M (r)smed,cav (r ) pQ (r )
Dcav
fmed
• smed,cav is the stopping power ratio medium to cavity
material
• pQ is an overall perturbation correction factor that
accounts for everything a cavity theory does
not account for
fmed is the ratio of dose to medium to dose to cavity;
factorization of fmed is questionable in regions of
strong disequilibrium
AAPM Summer School, 2006
12
Practical complications
• The outlined interpretation depends on our
ability to convert cavity dose to medium dose
– CPE or TCPE and detector is small compared to the
range of secondary electrons:
• fmed can be factorized in a stopping power ratio +
correction factors
– If these conditions are not fulfilled:
• Monte Carlo calculations (within the constraints of the
algorithm and basic cross-section datasets)
AAPM Summer School, 2006
13
Why is this important for commissioning
and accuracy verification of TPS?
For the measurement of a dose distribution:
• In regions of CPE and TCPE: SPR corrections
accurately represent detector response (and are
small for air-filled chambers in photon beams)
• In regions of non-CPE: SPR corrections DO NOT
accurately reflect changes in detector response and
additional, sometimes large, corrections are needed
– build-up regions in any field, interface-proximal points in
heterogeneous phantoms (build-up and build-down)
– narrow fields
– modulated fields
AAPM Summer School, 2006
14
Stopping power ratios (SPR’s)
• SPR’s are depth, field size, and radiation
quality dependent
– for reference dosimetry using ionization
chambers: values based on Monte Carlo
calculations and well documented
– for relative dosimetry in TCPE: their variation
relative to the reference point needs to be
established
AAPM Summer School, 2006
15
sw,air(z) for plane parallel bremsstrahlung spectra
60Co
10 MV
20 MV
30 MV
Andreo and Brahme, PMB, 31, 839 (1986)
Perturbation correction factors:
traditional factorization for ionization
chambers:
PQ = Pwall Pgr Pfl Pcel
• Pwall: wall perturbation
• Pgr: gradient effect due to displacement
of phantom material by cavity
• Pfl: fluence perturbation
• Pcel: central electrode perturbation
AAPM Summer School, 2006
17
Perturbation correction factors
• depth, field size, and radiation quality
dependent
– for reference dosimetry using ionization
chambers: based on Monte Carlo calculations
and relatively well documented
– for relative dosimetry: their variation relative
to the reference point is usually ignored but
can be very significant
AAPM Summer School, 2006
18
Ding and Wu, Med. Phys. 28, 298 (2001)
Buckley and Rogers, Med. Phys. 33 455 (2006)
Buckley and Rogers, Med. Phys. 33 1788 (2006)
Commissioning measurements
• beam model verification
– in-air measurements
– in-water measurements
• in-phantom verification
– heterogeneous measurements
• reference dose measurements / system
calibration
AAPM Summer School, 2006
22
In-air measurements
• Parameters of primary source determined
– mean energy of electrons exiting the vacuum
window
– FWHM of intensity distribution of electron beam
exiting the vacuum window
• Measurement issues?
– detector response can change as a function of off
axis distance
AAPM Summer School, 2006
23
Off axis profiles are
sensitive to the energy
of the primary electron
source.
Off axis profiles are
collision kerma (to
water or air) profiles.
Sheikh-Bagheri & Rogers
(2002) Med. Phys. 29, 379 390.
AAPM Summer School, 2006
24
Off axis profiles are
sensitive to the
radius of the
primary electron
source.
Build-up cap:
“hevimet”
Tonkopi et al (2005) Med Phys 32, 2918 - 2927
AAPM Summer School, 2006
25
In-air measurements:
Which build-up cap should be used?
Experimental validation
of different types of b.u.
caps.
Tonkopi et al (2005) Med Phys 32, 2918 - 2927
AAPM Summer School, 2006
26
In-air measurements:
Which build-up cap should be used?
Cap that approximates
collision air kerma the
best?
Note:
1. small difference
between two cap
sizes
2. -> resolution does
not significantly
decrease with large
cap size.
Tonkopi et al (2005) Med Phys 32, 2918 - 2927
AAPM Summer School, 2006
27
In-air measurements, wide beam
profiles for electrons
• in-air wide-field open beam profiles uncover
details about
– electron source parameters
– scattering foils
– filters in the beam
AAPM Summer School, 2006
28
Wide beam profile measurements for
Total Skin Electron Therapy
•
•
•
•
Total skin electron therapy is an external beam technique for treating mycosis fungoides. The
complicated set-up makes commissioning and treatment planning time consuming.
Monte Carlo electron beam modeling for large field size has poor agreement.
4 cm
In this work, we investigated the electron focal spot size by measuring with a slit camera.
The MC linac model uses the measured FWHM as the source parameter for the simulation.
0.4
Collimato
r 0.3 4 cm
In-air profile for 40 x 40 cm2 at 100 cm SSD
Jaws
Lucite
1.2 scattering
Primary
scatterer
foil at Z = 9.5 cm
Fit
Measurement
Measurement
0.2
Relative dose
Relative OD
1
20 cm
Electron beam0.1
0.8
Isocenter
Secondary scattering
0.6
foil at Z = 12.9 cm
0.4
0
20
24
-0.1
-40
Scattering
foils
MC
-30
-20
Ion chamber
lead (0.152
FWH 0.2 Alternating
32
mm)28and paper (0.0889
M
0 mm) sheets
-10
0
10
Relative distance (mm)
Off axis distance (cm)
Scattering filter
AAPM Summer School, 2006
20
30
40
Rotatin
g
platfor
29
m
Source description optimization
Focal spot is elliptical
and has a Gaussian
distribution.
FWHM in lateral direction
(mm)
2.40
2.10
1.80
Shift in lateral direction (mm)
Focal spot shift with respect to crosshair for various electron
Focal spot size for various electron energies
energies
3.0
6 MeV
9 MeV
12 MeV
16 MeV
1.5
2.11, 2.24
9.0
Lateral and
longitudinal shifts
with respect to the
crosshair is also
observed for all
measurements.
AAPM Summer School, 2006
30
7.7, 1.4
1.87, 2.14
1.69, 2.09
60.0
MeV
9 MeV
3.0
12 MeV
-1.5
16 MeV
1.50
-3.0
1.30
6.0
4.6, -0.9
6.5, -1.3
1.73, 1.75
5.7, -2.0
1.80
2.30
Shift in longitudinal direction (mm)
FWHM in longitudinal direction (mm)
Source description optimization
Comparison of measured and calculated in-air profile for 40 x 40
cm2 at 100 cm SSD
1.2
Measurement
initial MC model
1
N orm aliz ed v alue
proposed MC model
Proposed divergent
beam model shows
improved result for 40 x
40 cm2 in-air profile.
0.8
0.6
0.4
Z = 0 cm
0.2
Zvirtual = 5.92 cm
0
-40
-20
0
20
40
“Virtual source”
Distance from central axis (cm)
1.60
Beryllium exit window
at Z = 9 cm
0.086 mm
Primary scattering
foil at Z = 9.5 cm
1 mm
AAPM Summer School, 2006
31
In-water beam model verification
measurements
• In-water measurements are frequently used
for beam model optimization since:
– these measurements are performed as part of a
standard commissioning.
– reproduction of these measurements by the MC
planning system forms a core consistency test of
the system’s commissioning.
– reference dosimetry is carried out in water (e.g.,
TG-51 or IAEA TRS-398) and thus allows to link
the output measured to the source’s energy and
fluence parameters.
AAPM Summer School, 2006
32
Different measurement issues in
open photon fields in water
•
in-field, CPE or TCPE
– standard measurement, SPR is nearly constant and represents
detector response well; detector perturbations are small (point-ofmeasurement shift is 0.6*rinner upstream)
•
in-field, non-CPE (build-up)
– non standard measurement, SPR varies modestly but does not
represent variation in detector response well; detector perturbations
are large, detector dependent.
•
penumbra (non-CPE)
– detector size must be small relative to field-size and penumbra
width, SPR does not represent detector response well, lateral
fluence perturbations, detector dependent.
•
out-of-field (CPE or TCPE)
– detector size must be sufficiently large to collect sufficiently large
signal. SPE represents detector response well.
AAPM Summer School, 2006
33
In-water off-axis measurements:
Behaviour of avg. photon and avg. electron energy
full lines: photons
dashed lines: electrons
-> stopping power ratio
w/air is position independent
However: wall correction
changes by about 2%
since contributions to ionization
change for modest changes in
avg. photon energy.
Dohm et al (2005) Phys Med Biol 50: 1449 - 1457
AAPM Summer School, 2006
34
A14P ExRadin ionization chamber
Geometry of the measuring volume
Collecting
electrode
diameter: 1.5 mm
Separation: 1 mm
Effective volume of the chamber determined by electric
field in chamber.
Laplace equation: ∇ 2U = 0
Boundary condition: Surface potential
AAPM Summer School, 2006
35
1.5 mm field
ƒspecial
collimator
ƒDiameter of the field:
1 mm at 70 cm from
the source
AAPM Summer School, 2006
36
Narrow 1.5 mm field
Average energies at d=2.5 cm
4
Electrons
Photons
3
2
1
0
0
2
4
6
8
10
Off-axis distance (mm)
AAPM Summer School, 2006
37
Narrow 1.5 mm field
Fluence at d= 2.5 cm
normalized to the central axis value
1.2
Electrons
1.0
Photons
0.8
Dose profile
0.6
0.4
0.2
0.0
0
2
4
Off-axis distance (mm)
6
AAPM Summer School, 2006
8
38
Narrow 1.5 mm field
Dose water-to-air ratio
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1
0.9
0.8
0
2
4
6
8
Off-axis distance (mm)
AAPM Summer School, 2006
39
Narrow 1.5 mm field
OAR at d = 2.5 cm measured with A14P
1.2
A14P
1
A14P corrected
0.8
A14P corrected and
deconvolved
0.6
0.4
0.2
0
-8
-6
-4
-2
0
2
Off-axis distance (mm)
AAPM Summer School, 2006
4
6
8
40
Narrow 1.5 mm field
OAR at d=2.5 cm
1.2
HS film
1
Monte Carlo
0.8
A14P corrected and
deconvolved
0.6
0.4
0.2
0
-8
-6
-4
-2
0
2
4
6
8
Off-axis dstance (mm)
AAPM Summer School, 2006
41
Build-up dose measurements
• Build-up dose calculations are challenging
but may be clinically important for DVH’s of
superficial structures. However:
– Discrepancies have been reported between
measured and calculated build-up doses at high
photon energies.
– Benchmark measurements are needed to ensure
the system accurately calculates dose in the buildup region.
– Benchmark measurements are prone to
uncertainties ranging up to 50% of the local dose.
AAPM Summer School, 2006
42
Buildup region – 6 MV - 10x10 cm2
100
EGSnrc
NACP
PEREGRINE
80
Relative Dose (%)
60
13%
discrepancy
60
50
40
40
20
0
30
0.0
0.0
0.5
Depth in water (cm)
0.1
0.2
1.0
Buildup region – 6 MV- 40x40 cm2
100
NACP
EGSnrc
PEREGRINE
80
Relative Dose (%)
80
10% discrepancy
60
70
40
60
20
50
0
0.0
0.0
0.1
0.5
0.2
1.0
Depth in water (cm)
1.5
NACP chamber response in buildup region
⎛ Q ⎞⎛ W ⎞
Dcav = ⎜
⎟⎜ ⎟
⎝ mair ⎠ ⎝ e ⎠air
Dcav (d)
=
Q(d)
Dcav (dmax ) Q(dmax )
NACP
DOSRZnrc cavity
simulations
measurement
Field size
(cm2)
Calculated relative
surface cavity dose
Measured relative surface
ionization
10x10
45 ± 1 %
47 ± 2 %
40x40
67 ± 2 %
69.0 ± 0.3 %
AAPM Summer School, 2006
45
Perturbation correction factors – 10x10 cm2
⎛
⎜
⎜
⎜
⎝
Dwater (d ) = Dcav (d ) L(d ) ρ
⎞
⎟
⎟
⎟
⎠
water
air
water (d )
Φ air
Depth (cm)
Perturbation correction
0.0
0.78 ± 0.02
1.5
0.99 ± 0.008
AAPM Summer School, 2006
46
Sources of chamber perturbation
guard
ring
Effective point
of
measurement?
3 mm
5 mm
AAPM Summer School, 2006
47
Buildup region – 10x10 cm2
100
80
Relative Dose (%)
EGSnrc
60
PEREGRINE
40
PEREGRINE
overestimates dose
near surface by 6%
20
0
0.0
0.5
Depth in water (cm)
1.0
Reconciling
measurements
and
calculations in the
build-up region by
adjusting EPOM
->simple if it
works; but is
chamber
dependent!
Kawrakow (2006) Med. Phys. 33, 1829
AAPM Summer School, 2006
49
Measurements in the presence of
beam modifiers - MLC’s
• Leakage and Transmission
–
–
–
–
contributes >10% of open field dose in IMRT
spectrum (beam hardening)
contribution to organs at risk
widening of penumbra (rounded leaf ends)
• Tongue and groove effect
– 10-15% under-dosing for static fields
– no detailed study for IMRT
AAPM Summer School, 2006
50
Across leaf bank
100
IC04
DYNVMLC
Relative Dose (%)
80
60
40
20
0
-15
-10
-5
0
Inplane (cm)
5
10
15
Along direction of leaf motion
100
IC04
DYNVMLC
Relative Dose (%)
80
60
BEAM profiles
systematically
narrower (1-2
mm)
40
20
0
-15
-10
-5
0
Crossplane (cm)
5
10
15
Validation – 5 cm offset
inplane
100
DYNVMLC
IC04
crossplane
Relative Dose (%)
80
60
40
5 cm
20
0
-15
-10
-5
Crossplane (cm)
0
5
MLC bar pattern
+ Inplane
Relative Dose (normalized to open field)
100
80
DYNVMLC
Diode
60
40
20
0
-25
-20
-15
-10
-5
0
5
10
15
20
Inplane (cm)
AAPM Summer School, 2006
54
25
100
MC
CA24
Film
Rel. Dose (%)
80
60
40
20
0
-8
-6
-4
-2
0
2
4
6
8
Crossplane (cm)
100
CA24
MC
Film
Rel. Dose (%)
80
Static Delivery
60
40
20
0
-8
-6
-4
-2
0
Inplane (cm)
2
4
6
8
Importance of beam modeling for 3D-CRT
Monte Carlo – all tissues to water
CADplan
All tissues to water, plan 2
120
100
80
cord
60
The conventional planning
system uses a pencil beam
algorithm once
MLC is used as
beam shaping device
40
CADplan
xVMC
20
0
0
200
400
600
800
Dose (cGy)
1000
1200
1400
Validation – MLC leakage
Leaf material density and abutting leaf air gap were chosen to
match simulated MLC leakage profiles with film measurements
• leaf gaps measured
physically
• measured leakage is
sensitive to measurement
configuration
• simulation should match
measurement geometry
AAPM Summer School, 2006
58
Interleaf leakage at 2 cm offset
• density = 17.7
g/cm3
Film
Relative Dose normalized to Open field
• interleaf air gap =
0.0057 cm
2.0
BEAMnrc/DOSXYZnrc
1.5
1.0
0.5
0.0
-6
-4
-2
0
2
4
6
Inplane (cm)
AAPM Summer School, 2006
59
Interleaf leakage at 4 cm offset
0.2- 0.3 %
increase in
transmission due
to leaf driving
screw hole
2.0
Relative Dose normalized to Open field
•
1.5
BEAMnrc/DOSXYZnrc
1.0
Film
0.5
0.0
-6
-4
-2
0
2
4
6
Inplane (cm)
AAPM Summer School, 2006
60
Leakage between abutting leaves
• abutting leaf gap
= 0.004 cm
Relative dose normalized to open field
30
25
Film
BEAMnrc/DOSXYZnrc
20
15
10
5
0
-5
-4
-3
-2
-1
0
1
2
3
4
Crossplane (cm)
AAPM Summer School, 2006
61
5
Depth dose profiles
100
IC04
10x10
Relative Dose (%)
80
DYNVMLC
DYNVML
20x20
2x2
60
4x4
40
20
0
0
5
10
15
Depth in water (cm)
20
25
100
Rel. Dose (%)
80
MC
film
CA24
60
40
Film measurement
uncertainty ?
20
0
-8
-6
-4
-2
0
2
4
6
8
Crossplane (cm)
100
Rel. Dose (%)
80
Dynamic Delivery
MC
Film
CA24
60
40
20
0
-8
-6
-4
-2
0
Inplane (cm)
2
4
6
8
IMRT pattern
EDR2 d=1.5 cm ~200 cGy
(0.022 x 0.022 cm2 pixels)
BEAMnrc/DOSXYZnrc
(0.22 x 0.22 x 1 cm3
voxels)
IMRT pattern
rms diff = 3.5
%
Some conclusions on “Measurements in the
presence of beam modifiers”
• Use suitable detectors: photon diode, small volume
ionization chamber, diamond detector, radiochromic
film (EBT - humidity effects!)
• MLC leakage, transmission and field definition
calculations need to agree with measurements - think
about dose calculations to organs at risk!!
– measurements are used to optimize dimensions and
properties in calculation model
– the beam model must be realistic to ensure proper OAR
dose calculations ALSO for non-IMRT planning
– therefore: contact manufacturer if you are sure of your
measurements and agreement is not obtained
AAPM Summer School, 2006
66
In-phantom measurements to
verify dose calculation accuracy
• Assuming beam models have been verified and
found accurate, inherent dose calculation
accuracy in patient or phantom may be
compromised by implementation issues such as:
– voxel-to-region alignment, voxel sizes
– CT - interaction coefficient conversion
– runtime limitations, smoothing options
• Verification is needed in heterogeneous phantoms
– simple, slab-based phantoms
– more complex heterogeneous phantoms
AAPM Summer School, 2006
67
Slab Phantoms, photon beam, 6 MV
(Gammex -RMI, tissue equivalent materials):
1.0 cm
SW
3.0 cm
bone
1.0 cm SW
6.0 cm lung
20.0 cm
SW
20.0 cm SW
fluence
perturbation
of TLD in
narrow field?
Heath et al 2004
AAPM Summer School, 2006
69
Heath et al 2004
AAPM Summer School, 2006
70
TLD-700 response as a function of accumulated dose…
Slab Phantoms, electron beams
(Gammex -RMI, tissue equivalent materials):
1.0 cm
1.1 cm
SW
bone
3.0 cm SW
6.0 cm
SW
6.0 cm lung
9.0 cm SW
Doucet et al 2003
15 MeV electrons
Doucet et al 2003
9 MeV electrons
Pq (16, med )
= 1.014 0.4%
Pq (dmax ,SW )
Doucet et al 2003
15 MeV electrons
Doucet et al 2003
9 MeV electrons
Doucet et al 2003
Water phantom, Al rod on central axis, 15 MeV
Doucet et al 2003
Doucet et al 2003
More complex, realistic heterogeneous phantoms …
Layered lung-equivalent phantom (a) full slab and (b), tumor
geometry used in the measurement of dose. (Charland et al 2003)
AAPM Summer School, 2006
79
Direct verification of treatment plans, realistic phantoms
(a)
(b)
Bone-equivalent
hard
boneCheek
bone
equiv.
air-equiv.
example
detector
locations
Air –equivalent
sinus
Tissue –
tissue-equiv.
equivalent
tumor
~ 15 cm
tissue-equiv.
Air-equivalent
trachea
soft-bone
equiv.
Spinal cord
(a) sample detector locations within an anthropomorphic lung
phantom;
(b) phantom with interchangeable air, bone, and tissue
equivalent rods at illustrated locations, mimicking
head/neck tissues.
AAPM Summer School, 2006
80
Conclusions
•
•
•
•
•
Measurement issues are an important component of
commissioning and verification of a MC TP system.
The user of an MC-based TP system must be aware of
complications in measurements in non-equilibrium situations.
Suitability of detectors for a given purpose must be assessed
when planning system accuracy in these regions is at stake.
MC planning has the potential to be dosimetrically superior to
conventional planning systems especially in dose estimations
for OAR’s -> BUT: to this end, extra validation is needed.
Commissioning and verification should also carefully deal with
the other components of the implementation such as CT to
interaction coefficient conversion procedure, voxel statistics,
smoothing functions, etc all of which are specific to MC
planning systems.
AAPM Summer School, 2006
81
Acknowledgments
•
•
•
•
Students: Emily Heath, Kristin Stewart, Khalid Al-Yahya, Keyvan
Jabbari, Andrew Alexander, Adam Elliott
Post-doc: Gabriela Stroian
Staff: Frank Verhaegen, Wamied Abdel-Rahman, Slobodan
Devic, Ervin Podgorsak, Michael Evans, William Parker
Funding (operating): Canadian Inst. Health Research (CIHR),
National Cancer Inst. Canada (NCIC), Natural Sciences and
Engineering Research Council (NSERC)
AAPM Summer School, 2006
82