RDTH3120 Physique médicale 2

treatment planning

S. Vynckier
Cliniques univ. St-Luc, UCL,
Brussels, Belgium
GTV-CTV-PTV
Tumour GTV
CTV I CTV II
CTV I + CTV II PTV
Irradiated volume - Treated volume
Isodoses + BU zone
Combination of isodoses
Effect of wedges at 90°
Effect of wedges for missing tisue
2D treatment plans : normalisation points
TPS system
 TPS system

Stand alone system Multi-station system system

Dose calculation algorithms
 Dose calculation algorithms

„ Matrix representation
„ Separation of primary and scatter (Clarkson)
„ Pencil beam algorithms
„ Monte Carlo methods
Dose calculation algorithms:
matrix representation

matrix Decrement lines

 Dose calculation algorithms:
separation of primary and scatter
S(z,r) = T(z,r)-T(z,0)
D(x,y,z)= Dp(x,y,z) + Ds(x,y,z)
 Dose calculation algorithms:
pencil beams (photons)
„ Pencil Beam Algorithm :
„kernel : p(r,z)=(ρ/r) (A e z +B e z )
-a r -b r
z z
„D(x,y,z)=∫ ∫ ∫ ∫ (1/ρ).ψ (x,y,z).p(x,y,z) dxdydzdE
E x y z E
 Pencil Beam Convolution

D ( x , y , z ) = ∫ ∫ ∫ Ψ ( x ', y ', z )K z ( x − x ', y − y ' ) dxdy

From Colin Field, University of Alberta

 3D Convolution/Superposition

Source of
„ The dose distribution can be computed using a Radiation
convolution equation:

µ
D (r ) = ∫ Ψ (r ') K (r − r ')d r '
3
v
ρ r ′ rv
Primary Kernel Derived
Dose From Monte Carlo
Energy
Distribution Simulation
Fluence

()
(
K r − r' ) Dr
 Dose calculation algorithms:
Monte-Carlo methods
Monte Carlo or random sampling techniques are used to generate dose distributions
by following the histories of a large number of particles as they emerge from the
source of radiation and undergo multiple scattering interactions both inside and
outside the patient.

18 MV photons… 100 primary e- e- only e- and photons

 Methods of problem solving ...

Mathematical description available

Analytic Numerical Monte Carlo

formulation: math.expression algorithm statistical description

type of problem: simple complex very complex

precision: exact approximation statistical (mean, SD)

cpu time: none normal large

 area of unit circle (1) ...

Analytical (exact)
1 1
⎧1 x ⎫
A = 4 ∫
0
1 − x 2 dx = 4 ⎨ arcsin x +
⎩2 2
1− x2 ⎬ = π
⎭0
 area of unit circle (2) ...

1 2 π
Aout = r tan
2 n
1 2 2π
Ain = r sin
2 n

Numerical (approximate) π n 2π
n tan > A > sin
n 2 n
n= 4 4 2
8 3.313708 2.28427
16 3.182598 3.061467
1024 3.141603 3.141573
 area of unit circle (3) ...

MC (statistical)
N

∑f i
A = 4 i =1
= 4 f
N 1
⎧
⎪ ∑ fi −
2 (∑ i
f )2
⎫
⎪
2

⎪ N ⎪
SD = ⎨ ⎬
Choose xi, 0 ≤ xi < 1 ⎪ N −1 ⎪
⎪⎩ ⎪⎭
Choose yi, 0 ≤ yi < 1 N A SD
10 3.2000 0.5060
If xi2 + yi2 < 1 then ƒi = 1 1 000 3.0880 0.0531
1 000 000 3.1423 0.0016
else ƒi = 0 10 000 000 3.1417 0.0005
Why MC simulation of radiation transport?

Radiation transport is in general a very complex process

Measurements Analytical solution

often practical problems

• borders
• inside a detector • complex equations
• inside a patient (inhomogeneities, scatter)
• many exposure conditions
 MC simulation in practice…a single history

photon
1 2 3 electron

Compton
scattering Scoring
region
1 : sample particle 3 : sample type of
energy, direction, interaction
starting position, ...
2 : sample distance PP C PE
to interaction
4 : sample direction, Put them
Photon Energy energy, … of new on stack
Interaction distance particles
 A particle shower

9 MeV electrons water

100 5
N = 1x10
80
60
% Dose
40
20
0
0 1 2 3 4 5
Depth (cm)
100 6
N = 1x10
80
60
% Dose
40
20
0
0 1 2 3 4 5
Depth (cm)

100 6
N = 10x10
80
60
% Dose
40
20
0
0 1 2 3 4 5
Depth (cm)

100 6
N = 75x10
80
60
% Dose
40
20
0
0 1 2 3 4 5
Depth (cm)
Dose calculation algorithms:
Pencil beam, electrons

Fermi-Eyges theory
 Visualisation tools: isodoses

Isodose visualisation

Multi-planar (sagital)
Transversal plans plans
 Visualisation tools: DVH

PTV PTV

OAR Contour
Dose volume hystograms
DVH for PTV
DVH critical organs
DVH for total volume
Cranial treatments
Beam eye views
Beam eye views
Beam eye views +DRR’s
Isodose distributions
 Dose volume histograms of PTV

cumulative frequence
Dose volume histograms of OAR
Breast treatments
prostate treatment procedure
ORL conformal treatment
procedure
„ Disadvantages
„ poor knowledge of junction between photons and
electrons fields
„ Cold and hot spots

„ Poor sparing of parotids photons

„ Intended dose distribution electrons

underdosed area
nodal PTV according to Nowak et al. IJRO 45: 33-39, 1999
40 Gy and Wijers et al. R&0 52: 35-42, 1999

70 Gy
„
„ Same fields on 55 Gy
„ then conformal boost
„ (70 Gy)
„ ⇒ Dosimetric quality
control of this
technique
„ Comparison between
calculation by TPS and
measurements (in water
& in Alderson phantom)
„ Analysis of the technique
„ Advantages
„ no junction between
photons and
electrons fields
„ better sparing of
parotids and spinal
cord
„ better conformity
„ Same fields during
55 Gy
 Conformal treatment

Basic aspects of conformal radiotherapy

Target localization is achieved through anatomical and functional imaging: computed
tomography (CT), magnetic resonance imaging (MRI), single photon emission
computed tomography (SPECT), positron-emission computed tomography (PET

CT image MR image
 Patient anatomy identification

Identification of tumor will be achieved with modalities

as CT, MRI and PET and associated fusion techniques

CT MRI PET
 Conformal treatment

Basic aspects of conformal radiotherapy

Treatment planning is achieved either with standard "forward planning" techniques,
shaping the blocks or MLC to the target volume

block MLC
 Conformal treatment

Basic aspects of conformal radiotherapy

Treatment planning is achieved either with standard "forward planning" techniques,
shaping the blocks or MLC to the target volume

Multi-planar (sagital) plans

Transversal plans
 Conformal treatment

Basic aspects of conformal radiotherapy

Dose delivery techniques range from the use of standard regular and uniform coplanar
beams with conformed blocks or with multileaf collimators (MLCs).
 Conformal treatment
QA on delivery techniques: blocks
 Conformal treatment
QA on delivery techniques: MLC’s
Mechanical checks: MLC
Mechanical checks MLC’s : recording
 Concave dose distribution component:
intensity for isotropic convergent coplanar beams

Intensity part to compensate:

- extra tissue (1)
- shadowing of other beams by OAR (2)
2
1
 Concave dose distribution component:
intensity for isotropic convergent coplanar beams

Homogeneous
beam Modulated beam
 Protection of the rectum: DVH
(2D)

3D simple

3D
segmented

Gain dose-volume
 Géométrie du faisceau

Optimisation des intensités

IMRT
Distributions de
Fluence
dose
optimisée

Séquenceur
livraison

Fluence
délivrée
 IMRT
“conventional” conformation intensity modulation

treated volume

target volume
risk organ
 Intensity Modulation:
wedge filters, compensators

Wedge filter

Tin – Wax Compensators

Beam Weight Distribution
 Multileaf Collimators (MLC’s)
Conventional Binary
Siemens Varian NOMOS

• Conventional MLC’s were designed for field shaping and have limitations when
used for IMRT.

• Binary (off-on) MLC’s are designed for IMRT and are the easiest to model and
verify.
 Step & Shoot IMRT

1/3
Dose
1. mMLC field Set-Up

1/3
2/3
Dose Dose

2. mMLC field Set-Up Total Resulting

Dose

1/3
Dose
3. mMLC field Set-Up
 How?
Méthode step and shoot
Cummulative % of MUs
Segment #

„ Épargne moelle 1 6.
„ épargne parotides
„ 15 segments
„ 10 niveaux
IMRT: Dynamic MLC (DMLC)
Continuous irradiation
 Segmental MLC IMRT
vs.
Dynamic MLC IMRT
SMLC DMLC
„ Only discrete intensity levels „ Continuous intensity levels.
can be delivered.
„ Pairs of MLC leaves are in
„ Deliver multiple MLC apertures continuous movement across the
within a field to apply the field with the intensity at a point
intensity in a paint-by-number equal to the total exposure time
fashion. of the leaf pair above it.

„ May be a straightforward „ Most efficient delivery for modest

technique for “forward” modulation of intensities.
optimization.
„ High spatial variation of
„ The whole intensity map and intensities are difficult.
each segment can be verified
using portal imaging.
„ Difficult to verify with
„ May be relatively time conventional techniques;
consuming if field delivery is mouvement of the leaves is part
verified in exactly the same of the QA process.
way.
 Conversion from Continuous to
Discrete Intensity Profiles: segmentation

DMLC IMRT SMLC IMRT

From Chen Chui, Memorial Sloan Kettering

 Intensity Modulated Arc Therapy (IMAT)

„ Collimator leaves move dynamically as the gantry rotates.

„ Beams delivered from all coplanar directions.

„ Requires multiple arc deliveries.

„ Provided field length is not too long, no couch translations

are necessary.

„ Proposed by Cedric Yu and implemented in Gent, Belgium.

 Robotic Pencil Beam Delivery
„ The most degrees of freedom.

„ No discretization of beam intensity levels is

necessary.

„ Lowest beam efficiency.

„ Very slow to treat a large volume.

„ Stationary x-ray beams for image guidance.

„ Investigations of its use in IMRT are only beginning.

Robotic Pencil Beam Delivery using the Accuray
Cyberknife™

From Ken Gall

Conformal radiotherapy and IMRT
 Immobilization of Head &
Neck patients

z 15 head & neck tumors

z 15 brain tumors
z 3 different masks (3, 4 & 5 FP)
z 3 fictitious isocenters
 Immobilization of Head & Neck patients

z 915 portals images acquired.

z 98 % analysed.

z Total Displacement.
‹ SIM - EPIDi

1.
Cumulative incidence

0.
0
90.8 Head
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 1 2 3 4 5 6 7 8 9 10
Deviation from the isocenter (mm)
 Methods for
Pre-Treatment Setup Verification

„ Rigid Immobilization
– Stereotactic systems for head or body

„ 2D+Time Portal Verification

– Electronic portal imaging systems and implanted fiducials

„ Ultrasound Guidance
– Pelvic ultrasound to visualize the prostate

„ CT Guidance
– In-Room kV
– In-Room MV
 Why 3D Image-Guided Radiotherapy
(IGRT)?
„ Eventually a large number of radiotherapy treatments will be
IMRT.

„ All IMRT should be image-guided:

– IMRT is justified by sparing critical tissues (conformal avoidance)
which produces higher dose gradients.
– IGRT enables higher gradients to be delivered safely and effectively.
– IGRT enables a smaller PTV margin to be defined.

„ In some radiotherapy sites, (e.g., prostate movements),

IGRT may be more important than IMRT.

„ 2D imaging is inadequate to obtain volume information.

 In-Room Kilovoltage CT

Helical KV CT:
Separate KV CT
and Linac

Siemens, GE

KV CT and Linac
On Same Gantry

Elekta Synergy
VarianTrilogy
Siemens
 Cone beam CT

Accelerator-based Acquisition; 320 Projections; 120 kVp, 200 mAs; 180 s.

(0.25 x 0.25 x 0.25) mm3 voxels (courtesy, NKI)
 Helical Tomotherapy: …

Designed for IGRT & IMRT:

integration!
 Tomotherapy

„ Système Hélicoidal
„ L’anneau est un scanner modifié
Tunnel Î 85 cm
„ “fan beam “ de 6MV
„ Faisceau modulé par 64 lames CML
„ Vérification de la position du patient
en temps réel par système CT
Tomotherapy 6MV High Dose Source
(800MU/min, 1.5mm point source)

Primary Collimator
(0 to 5.0 cm)
Binary MLC
(64 leaves, ea @ 0.61cm)
85 cm Gantry Aperture
Approximate 85cm
40 cm VRCT FOV

Approximate 50cm

VRCT Detector System

Helical tomotherapy, under the covers

Gun Board Linac

Control
Computer

Circulator

Magnetron

Pulse Forming
Network and
Modulator

Data Acquisition System

High Voltage
Power Supply Beam Stop Detector
Integration Around:Scientific Computer System

Tomotherapy
Unit

Planning Operator
Station Station
Common
Database and Cluster
Tomotherapy
Helical Tomotherapy: …
Total Marrow
Irradiation
(TMI) Using
Tomotherapy

Conformal
Avoidance of:
Brain
Thyroid
Lungs
Liver
Kidneys
Small Bowel

From Dr. An Liu, City of Hope, Duarte CA

Planning Station
Register Verification CT to Planning CT

Auto Man
Register Verification CT to Planning CT
Register Verification CT to Planning CT
 Comparison of Tomo with CBCT
Density Plugs +3% Contrast
Water
-6% Contrast
Tomo Synergy

1 cGy 3 cGy

UW Madison Tomotherapy Unit UC Davis Synergy Unit

 MVCT: high Z material
Double Hip prosthesis : kVCT Double Hip prosthesis : MVCT
 QA of conformal therapy / IMRT

Calculated Patient
fluence model

Commissioning
Calculated dose

Machine QA ? Patient QA

Delivered dose
Pre-treatment QA
Delivered Real
fluence patient
 QA of Conformal treatment and IMRT

„ Description of the QC
– in water
„ pdd and profile
„ output
– “in vivo” with Alderson phantom
„ CT-Scan of an Alderson phantom
„ Treatment plan on this phantom contoured by MD
„ Simulation of the TP fields and export of data to linac
„ Irradiation of phantom
„ Measurement with TLDs
„ Calculation with TPS
 Conformal radiotherapy and IMRT

Image acquisition
 Conformal radiotherapy and IMRT

QA: Comparison between

calculation by TPS and
measurements (in water
& in Alderson phantom)

Courtesy of M. Tomsej
 P DD me asu red difference

Results of the QC
PDD Hela x

„ 1.2 1

– In water
0.5
1

„ Relative PD
D
0.8
-0.5
diff
ere
nce
(%)

– pdd (max 2% at 25 0.6

-1

cm H20 in depth) 0.4

-1.5

– profile at 5cm H20

-2

(max 1%)
0.2 -2.5

0 5 10 15 20 25 30
dep th cm H2O

„ output
– Difference: 2.0 %
at 5 cm H20 125

p ro file z = 5 c m TM S
m e a su r e d
100
r a tio

75
rel
a ti
ve
dos
e + 50
rat
io
(%
)
25

-2 5
-7 .5 -5 -2 .5 0 2 .5 5 7 .5
p o sitio n (c m )
Conformal radiotherapy and IMRT

In Alderson phantom

Point dose
10.0
9.0
8.0
7.0
6.0
+ 5%

TMS / TLD (%)

5.0
4.0
3.0
2.0
1.0
0.0
-1.0
-2.0 0 2 4 6 8 10 12 14 16 18 20
-3.0
-4.0
-5.0
-6.0
-7.0 - 5%
-8.0
-9.0
-10.0

TLD number

Courtesy of M. Tomsej
 QA of IMRT: Linac QA

„ Beam quality: photon beam 6MV, SSD=100

cm
„ Field sizes: 2*2 cm2, 4*4 cm2 and 6*6 cm2
„ Field properties: open MLC fields
„ symmetrical fields (2*2 cm2, 4*4 cm2 and 6*6 cm2)

„ asymmetrical fields
– “offset” fields (2*2 cm2, 4*4 cm2 and 6*6 cm2)

– “half-fields” (2*2 cm2, 4*4 cm2 and 6*6 cm2) (ELEKTA)

QA of IMRT : patient QA

100

relative dose (%)

80

60

40

20

0
-10 -5 0 5 10
off-axis position (cm)
 QA of IMRT: Patient QA

•Relative dose verification

•Absolute dose verification