Eclipse eMC

Question 1

2 models in eMC algorithm

Answer 1

-model to transport electrons
-model to describe electrons and photons emerging from the head

Question 2

transport model

Answer 2

local to global MC
-MC performed in local geometry- get library of PDFs for relevant geometries and energies (uses EGSnrc code)
-MC calcs done in global geometry- Particles are transported through the CT
volume in macroscopic steps based on the PDFs generated in the local calculation

Question 3

how is CT volume pre-processed

Answer 3

To each voxel of the density-volume, a sphere index is
assigned that corresponds to the maximum sphere radius that can be used from the current voxel
center without the corresponding sphere reaching into the other material.

● Small spheres assigned to voxels located near interfaces between materials
● Large spheres assigned to voxels at greater distances from material interfaces (see the
figure)

Question 4

2 types of smoothing mechanisms in eclipse e MC

Answer 4

Gaussian- convolution with Gaussian
median smoothing - If a pixel is considered representative of its surroundings, it is replaced with
the median of the pixel values in the neighborhood

Question 5

what can be configured in eMC?

Answer 5

-statistical uncertainty
-calculation resolution
-random generator sequence used
-Number of particle histories: Defines the number of particles used in in the simulation. 0
value means that this option is not used; instead the simulation uses as many particles as
required to reach the statistical uncertainty set in Statistical uncertainty option.
-smoothing method and level

Question 6

open field measurements required for eMC

Answer 6

40x40cm2, no applicator
-dept dose in water at SSD 100 cm
-One profile in air at 95 cm. The measurement must extend at least up to a distance,
which corresponds to the diagonal length of the largest applicator size. The measurement
should contain only the contribution of the electrons.

Question 7

applicator measurements required for eMC

Answer 7

● Depth-dose curve in water at the Source-to-Phantom Distance (SPD) = 100 cm
● Absolute dose in water, expressed in [cGy/MU], at the calibration point on the depth dose
curve close to dref (dref = 0.6 R50 − 0.1 cm, where R50 is the point where the dose is
50% of the dmax).
● Inplane (gantry-table direction) and crossplane (perpendicular to inplane) profiles in air at
95 cm with jaws set to the positions they would be during the beam delivery with the
applicator, but without the applicator in place.

Question 8

eMC beam data created based on measurements

Answer 8

-open field PDD- to get energy spectrum of electrons from primary source
-applicator PDD- to get energy spectrum of electrons from jaws source
-Applicator absolute dosimetry
-Open field air profiles- construct 2D electron fluence
-applicator air profiles- optional- used for 2D electron fluence, \

The applicator air profiles are optional inputs for configuration. If no applicator air profile
measurements are provided, then radially symmetric fluence f (r) from the open field air profile is
used during dose calculation.

Question 9

2 models in algorithm

Answer 9

-source model
-transport model

Question 10

what structures are included in the algorithm model?

Answer 10

-collimator jaws
-MLC is modeled only indirectly as a part of jaw modeling
on the machines that have fixed jaws or no jaws parallel with the MLC
-blocks
-tray transmission not considered
-support structures (ex couches) are ignored unless inside the body

Question 11

electrons that emerge from spheres

Answer 11

If more than one electron emerges from
a sphere, the electron with the highest energy is called the primary electron. The other particles
are called secondary particles (secondary electrons and Bremsstrahlung photons).

-remember these spheres are all previously calculated for particular sizes and materials

Question 12

sphere materials and sizes in library

Answer 12

-air, lung, water, lucite, solid bone
-diameter of 1,2,3,4 and 6 mm
The maximum sphere
size depends on the particle’s initial energy: 2 mm for 4MeV, 3 mm for 5MeV, 4 mm for
6MeV, 6 mm for 7,5MeV and higher energies.
● 30 incident energy values Ti
(0.2, 0.4, 0.6, 0.8, 1, 1.5, 2, 3, … 24, 25 MeV)

Question 13

what information is stored for each sphere/energy/size combination from the library?

Answer 13

● Exit energy T f
, exit position α and exit direction θ of the primary electrons.
● Energy, direction and probability (weight) of secondary electrons.
● Energy and probability (weight) of secondary photons.

Question 14

what does eMC do with a material that has intermediate density between 2 materials?

Answer 14

assigns it a probability of being one material vs the other

Question 15

what makes a region considered homogeneous in eMC?

Answer 15

density of a voxel vs its neighbour is within 1.5

-If the densities in both voxels are
below the threshold of 0.05 g/cm3
, the ratio is not evaluated

-Entering a density threshold
prevents noise in low-density data from being interpreted as heterogeneity. For densities and
density ratios below the limits mentioned above, the MMC algorithm is capable of processing
differences in the material without decreasing the step size.

Question 16

what happens if mass density of a material exceeds the max mass density in the eMC database?

Answer 16

maximum mass density will be used and scattering may not be based
on the correct material.

Question 17

where is the sphere placed?

Answer 17

-The current position of the primary electron is the exit
position on the previous sphere.
-sphere center is placed at one radius of the max allowed sphere size of previous step

Question 18

when does eMC stop the particle at an interface?

Answer 18

-ratio of collisional stopping powers > 1.5

Question 19

how is energy deposited along each voxel along a ray?

Answer 19

Tvox = Tdep (Svox/Skug)(tvox/tdep)
tvox= length of line in voxel
tdep= total length of current transport step
Tdep= primary electron energy to be deposited in current step
Svox = linear stopping power of the voxel material
Skug=Linear stopping power of the sphere that was used to look up the resulting
electron parameters of the current transport step

-therefore, even though a sphere might be considered homogeneous, still looks at difference in stopping powers of voxels within the sphere

Question 20

how is primary photon treated?

Answer 20

The
direction of the primary photon may change in these interactions. Distance between the
subsequent primary photon interactions is chosen randomly based on the photon’s mean free
path and the attenuation coefficient at the given density and the photon’s remaining energy. The
primary photon is traced until it has lost all its energy.

Question 21

how is secondary electron treated?

Answer 21

The energy and the direction of the secondary electrons are sampled from the
distributions obtained from the local simulations.

Question 22

how are secondary photons treated?

Answer 22

only the energy is
sampled, direction is taken from the incident electron

-the photon is tracked as an electron- For the secondary photon, the distance to the next interaction is determined by taking the local
mass attenuation of the material into account and applying ray tracing from the sampled starting
position along the direction of the incoming primary electron. At the interaction position, the
energy which is transferred from the photon to the electron is sampled from a PDF depending on
the photon energy using data from XCOM.39 The weight of the electron is set to the weight of the
photon. The photon is no longer tracked.

Question 23

how is energy to each voxel calculated for secondary electron?

Answer 23

E = wTi x 2 MeV/cm * l * (Svox/Swater)
w(Ti) = Weight of the secondary particle emerging from incident electron of energy
Ti
. Δl = Length of the line inside the voxel.
Svox = Linear stopping power of the voxel material.
Swater = Linear stopping power of water.

i.e. assumes medium is water and multiplies by corection factor for stopping power of voxel compared to water

Question 24

what is w wrt to secondary electron?

Answer 24

probability that the secondary
particle occurs in the local simulation of the incident electron of energy Ti

Question 25

statistical uncertainty

Answer 25

refers to dose variation about a point, not necessarily to wether that point is accurate or not

Question 26

gaussian smoothing

Answer 26

produces a weighted average of each pixel’s neighborhood, with the
average weighted more towards the value of the central pixels

smoothing based on mean

Question 27

median smoothing

Answer 27

The median dose smoothing determines the value of a pixel by examining the pixel values in its
neighborhood on a slice and taking the median of these values

Question 28

what can user set in eMC?

Answer 28

-statistical uncertainty
-calculation resolution
-RNG seed number
-number of particle histories. If 0, just uses as many histories as needed to achieve uncertainty
-smoothing method and level

Question 29

sub-sources in electron source model

Answer 29

● Primary source (electrons and photons) as a point source near the scattering foil
● Jaws source (electrons and photons)
● Scraper sources for electrons scattered at the upper applicator scrapers
● Edge source for electrons scattered at the last applicator scraper or insert
● Transmission photons

Question 30

primary souce

Answer 30

90 cm SSD (ie 10 cm down from source)
- particles are sampled
on a plane located 95 cm below the nominal target, inside the shape defined by the applicator or
the insert.

● The direction of the photons is given by the sampled position on the plane and the Zposition of the focus (10 cm below the nominal target).
● The mean direction of the electrons is given by the sampled position in the plane and the
position of the nominal target. The direction is then varied according to a Gaussian
distribution with an energy dependent variance

Question 31

jaw source

Answer 31

particles are sampled on a
plane at 95 cm below the nominal target inside the shape defined by the applicator or the insert.

● The direction of the photons is given by the sampled position in the plane and the
sampled origin position.
● The mean direction of the electrons is given by the sampled position in the plane and the
sampled origin position. The direction is then varied according to a Gaussian distribution
with an energy dependent variance.

Question 32

scraper sources/edge electrons

Answer 32

The electrons are sampled from pre-calculated scatter kernels.

Question 33

transmission photons

Answer 33

sampled using pre-calculated kernels. They exit from the outer rim of
the applicator and from the insert material.

The following three types of transmission photons are taken into account:
● Scattered photons produced by main electrons in the insert material, sampled by using a
pre-calculated kernel.
● Photons passing through the insert material without interaction. They have the same
direction as the main photons but a different energy distribution.
● Scattered photons produced by main photons in the insert material, sampled by using a
pre-calculated kernel.

Question 34

normalization in eMC

Answer 34

-issues with normalizing to a single pt due to noise (uncertainty of a point is translated to uncertainty in normalization)
-better to normalize to a volume where uncertainty is averaged out
-can use smoothed dose max

Question 35

known limitations of eMC

Answer 35

-unusual contours, heterogeneities
-asymmetric cutouts
-dose near applicator borders
-outside field at shallow deptns
-elekta 6x6 applicators

Question 36

why do we have the in-air measurements?

Answer 36

-open field one is used to make phase space for primary source
-applicator in-air measurements (without applicator in place) are used to fine tune the phase space for each applicator (i.e 2D electron fluence)
-then only have to sample cutout

Question 37

where does bias come from?

Answer 37

-approximations
-RED curve errors

Question 38

why does eMC use 95 cm for measurements?

Answer 38

-plane of electron cutout

Question 39

what detector to use for electron in-air measurement?

Answer 39

small, with minimal buildup
-want to get dose from electrons only

literature: electron diode without build-up cap

Question 40

free MC software

Answer 40

PRIMO

Question 41

how to verify electron plans

Answer 41

-rad calc
-film
-spot checks with mosfets
-ion chamber array or diode array set up for electrons

Question 42

when is applicator in place?

Answer 42

-40x40 open field and water depth dose measurements- no applicator in place
-PDD curvs in water ad absolute dose in water at calibration pt- applicator in place
-inplane and crossplane proiles in air at 95 cm with jaw positions set but no applicatior in place

Question 43

why was uncertainty and smoothing level different depending on energy?

Answer 43

-balance accuracy with clinical time constraints

Question 44

how to validate cutout factors

Answer 44

compare eMC to measured on machine