Experimental validation of Monte Carlo simulation of Leksell ......of the Gamma Knife Perfexion stereotactic radiosurgery system performed using the Fluka code. The LGP PFX (Leksell

UNIVERSITÀ DEGLI STUDI DI MILANO

Facoltà di Scienze Matematiche, Fisiche e Naturali

Corso di Laurea in Fisica

Experimental validation of Monte Carlo simulation of Leksell Gamma

Knife Perfexion stereotactic radiosurgery system

Relatore interno: Prof.ssa Flavia Groppi

Relatore esterno: Dott. Giuseppe Battistoni

Correlatore: Dott. Maria Grazia Brambilla

Tesi di Laurea di:

Nicola Bertolino

Matricola 571204

Codice P.A.C.S. 87.53.-j

Anno Accademico 2008-2009

Riassunto

L'obiettivo del lavoro di questa tesi è la validazione sperimentale della simu-

lazione Monte Carlo con il codice FLUKA della testata per la radiochirurgia stere-

tassica Gamma Knife Perfexion della Elekta.

La radiochirurgia stereotassica si basa su interventi radioterapici di precisione

in cui viene utilizzato un sistema di riferimanto cartesiano per irraggiare la zona

interessata, minimizzanto la dose impartita alle zone cicostanti.

Il sistema Gamma Knife Perfexion è costituito da un'unità radiante dotata di

192 sorgenti di cobalto-60 opportunamente collimate verso un punto focale. Sono

presenti collimatori di tre dimensioni di�erenti (16, 8, 4 mm di raggio). Il sistema

Gamma Knife è utilizzato per radioterapia nella zona cerebrale. Per �ssare la testa

del paziente al letto del GK viene avvitato un casco al cranio del paziente e tramite

una CT (Computer Tomography) inserita in un opportuno software di Elekta viene

eseguita la piani�cazione del trattamento.

Il sistema informatico di Elekta per la piani�cazione dei trattamenti con Gamma

Knife, LGP PFX (Leksell Gamma Plan Perfexion), non tiene conto di cavità, diso-

mogeneità e zone di interfaccia. Infatti, l'algoritmo utilizzato da LGP PFX è basato

su un modello sempli�cato di interazione delle radiazioni gamma con la materia, che

considera omogeneo il mezzo utilizzato. Il metodo Monte Carlo potrebbe risultare

uno strumento utile per a�rontare questa problematica. Da una tesi collegata è

stato sviluppato il primo tentativo di simulazione completa del citato Gamma Knife

con il codice FLUKA.

La parte sperimentale di questo studio è stata eseguita al centro Gamma Knife

dell'ospedale Niguarda Ca' Granda'.

La prima prova è stata una misurazione della dose impartita all'isocentro ef-

fettuata tramite l'utilizzo di una camera a ionizzazione ad aria libera, 31016-PTW

Freiburg, con un piccolo volume sensibile inserita in un fantoccio composto da Gam-

mex 457 SolidWater (Leksell Gamma Knife Dosimetry Phantom) avente un diametro

i

di 160 mm, che modellizza una testa umana standard. La camera a ionizzazione è

stata introdotta dentro il fantoccio, che è stato posizionato in modo tale che la stessa

risultasse nel punto focale dei raggi gamma del Cobalto-60 dell'unità radiante. E'

stata quindi e�ettuata una misurazione con tutti i collimatori da 16 mm aperti,

i cui risultati sono stati successivamente comparati con quelli ottenuti tramite la

simulazione FLUKA eseguita riproducendo le medesime condizioni. L'errore nella

simulazione Monte Carlo (5 %) è dominato dall'incertezza sull'attività delle sorgenti,

mentre l'incertezza sperimentale calcolata è del 3 %. L'accordo fra i due valori risulta

all'interno dei margini di errore. Questa misura sperimentale è particolarmente im-

portante poichè è l'unico valore richiesto come informazione esterna dal software

LGP PFX.

Successivamente è stato e�ettuato un confronto della dose impartita all'isocentro

con le simulazioni per diverse aperture dei collimatori con i valori sperimentali.

Questa è anche una prima veri�ca del sistema computer-assistito per la piani�cazione

dei trattamenti LGP, infatti la dose sperimentale è ottenuta moltiplicando il valore

di dose sperimentale precedentemente ottenuto con camera a ionizzazione per gli

Output Factor (OF) forniti da Elekta. Gli OF sono il rapporto fra la dose impartita

da un set di collimatori (8, 4 mm) e la dose impartita da i collimatori da 16 mm.

Anche in questo caso risulta un buon accordo.

L'ultima veri�ca sperimentale realizzata in questo lavoro è un confronto fra pro�li

di dose relativa. I pro�li di dose sono misurazioni della dose trasversale nei piani

perpendicolari ai fasci di radiadione. Per queste misurazioni sono state utilizzate

pellicole radiocromiche EBT-ISP. Queste pellicole sono state preparate e posizionate

all'interno del fantoccio con l'ausilio di appositi supporti ed esposte a dosi crescenti

da 0 a 10 Gy in modo da ottenere una curva di calibrazione, che corregge la risposta

nonlineare alla radiazione delle pellicole radiocromiche. E�ettuata la calibrazione,

sono state eseguite tre serie di misure per ogni con�gurazione di collimatori (16, 8, 4

mm), impartendo un dose pari a 5 Gy. Le pellicole sono state quindi analizzate per

mezzo di un procedimento digitalizzato mediante uno scanner (Epson Expression

10000XL) per un confronto gra�co con i valori ottenuti dalla simulazione Monte

Carlo, ottenendo un buon accordo.

Dopo queste valutazioni sperimentali, le simulazioni basate sul codice Monte

Carlo FLUKA risultano convalidate. Un ra�namento delle simulazioni, con una

maggiore precisione della geometria e una statistica più grande, e delle misure sper-

imentali, riducendo l'errore statistico e aumentando la risoluzione della scanner, è

ii

prevista nello sviluppo di questo studio.

Dopo questa dovuta convalida del codice FLUKA è previsto uno studio appro-

fondito della sottistima e della sovrastima della dose impartita dovuta all'algoritmo

del software LGP PFX. E' previsto l'utilizzo di un fantoccio modi�cato con inserti

di materiali più densi o di cavità. Sarà poi possibile eseguire TC (Tomogra�e com-

puterizzate) del fantoccio in modo da piani�care trattamenti tramite il software di

Elekta. Con nuove misure sperimentali eseguite utilizzando la nuova con�gurazione

del fantoccio e con simulazioni che riproducono la situazione sperimentali si può

testare il piano di trattamneto del LGP.

iii

Contents

Abstract 2

1 Radiation therapy 5

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2 Stereotactic radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 Gamma Knife 10

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.2 Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.3 Gamma Knife Perfexion model . . . . . . . . . . . . . . . . . . . . . 13

2.4 Leksell Gamma Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3 Isocentre dose measure with ionization chamber 19

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.2 Gamma Knife Perfexion Dosimetry Phantom . . . . . . . . . . . . . . 20

3.2.1 Dosimetry Phantom With Ionization Chamber . . . . . . . . . 21

3.3 Ionization Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3.3.1 Determination of the absorbed dose . . . . . . . . . . . . . . . 24

3.4 The Bragg-Gray Cavity Theory . . . . . . . . . . . . . . . . . . . . . 25

3.4.1 Bragg-Grey Theory and Ionization Chamber . . . . . . . . . . 27

3.5 Experimental procedure . . . . . . . . . . . . . . . . . . . . . . . . . 28

3.6 Data discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.6.1 Output factor . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4 Relative dose pro�les with radiochromic EBT 31

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

4.2 Radiochromic EBT �lms . . . . . . . . . . . . . . . . . . . . . . . . . 32

4.2.1 Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

1

CONTENTS CONTENTS

4.2.2 Physical and chemical behavior . . . . . . . . . . . . . . . . . 34

4.3 Leksell Gamma Knife Dosimetry Phantom �lm stack . . . . . . . . . 36

4.4 Experimental Procedure . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.4.1 Cutting marking and positioning . . . . . . . . . . . . . . . . 37

4.4.2 Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.4.3 Scan and analysis . . . . . . . . . . . . . . . . . . . . . . . . . 38

4.4.4 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4.5 Data discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

4.5.1 Error in the Gafchromic EBT evaluated dose . . . . . . . . . . 40

4.5.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Conclusion 56

Bibliography 57

2

Abstract

The object of this work is to verify experimentally the Monte Carlo simulation

of the Gamma Knife Perfexion stereotactic radiosurgery system performed using the

Fluka code. The LGP PFX (Leksell Gamma Plan PERFEXION), an Elekta infor-

matic system for the planning of the Gamma Knife treatments, does not consider

interface zones and inhomogeneities. The algorithm improved in the LGP PFX is

based in a simplify γ ray interaction model with matter, considering the crossing

media like homogeneous (1, 17, 18, 19). Monte Carlo method (and FLUKA in

particular) could be a valid instrument to face this problem.

The experimental part of the study has been carried out in the Niguarda Ca'

Granda hospital Gamma Knife center.

The �rst test was a isocentre dose measurement performed by means of a free-

air ion chamber, 31016-PTW Freiburg with a small sensible volume, inserted in a

phantom made of Gammex 457 SolidWater (Elekta Dosimetry Phantom) with a

diameter of 160 mm, simulating a simplify standard human head. The ion chamber

was placed in the center of the phantom placed at the 60Co rays focus point of the GK

unit. A dose measurement with all the 16 mm collimators open was performed and

compared with the results of the FLUKA simulation in the same conditions. The

error in the FLUKA simulation (5 %) is determined by uncertainty in the source

activity, while the estimated experimental error is 3 %. The agreement between

the two values is inside the margin of error. This experimental measurement is very

important, because it is the only value requested as input by the LGP PFX software.

As second step a comparison between the simulated 4 and 8 mm collimators'

incendiary dose and the experimental value was made with a good agreement.

The �nal veri�cation included in this study is a comparison of the relative dose

pro�les. For this measure radiochromic EBT-ISP �lms were used. Gafchromic �lm

was prepared and placed inside the phantom by apposite holder and exposed to a

rising dose from 0 to 10 Gy in order to obtain a calibration curve to correct the non

3

CONTENTS CONTENTS

linear response of the �lm to radiation dose. After the calibration run, three set

of measurement in the coronal and axial plane were performed for all collimators'

con�guration (16, 8, 4 mm), for a 5 Gy dose. The �lms were scanned with a scanner

(Epson Expression 10000XL) for a graphical comparison with the MC simulation

values. Also in this case a good agreement between experiment and simulation was

achieved.

After these experimental evaluations, the simulation based on the FLUKAMonte

Carlo Code appear to be a valid instrument to test the LGP software and for future

studies about non homogeneous targets.

4

Chapter 1

Radiation therapy

Contents

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2 Stereotactic radiosurgery . . . . . . . . . . . . . . . . . . . 7

1.1 Introduction

Radiation therapy (also radiotherapy or radiation oncology, sometimes abbrevi-

ated to XRT) is the medical use of ionizing radiation as part of neoplastic pathologies

treatment and some not neoplastic pathologies. Radiation therapy works because

the radiation destroys the cancer cells ability to reproduce and the body naturally

gets rid of these cells.

Radiotherapy may be used for curative or adjuvant cancer treatment. It is used

as palliative treatment (where cure is not possible and the aim is for local disease

control or symptomatic relief) or as therapeutic treatment (where the therapy has

survival bene�t and it can be curative). Radiotherapy has several applications in

non-malignant conditions, such as the treatment of trigeminal neuralgia, severe thy-

roid eye disease, pterygium, pigmented villonodular synovitis, prevention of keloid

scar growth, and prevention of heterotopic ossi�cation. The use of radiotherapy in

non-malignant conditions is limited partly by worries about the risk of radiation-

induced cancers (9).

Radiotherapy is used for the treatment of malignant tumors may be used as

the primary therapy. It is also common to combine radiotherapy with surgery,

5

Chapter1. Introduction

chemotherapy, hormone therapy or some mixture of the three. Most common can-

cer types can be treated with radiotherapy in some way. The precise treatment

intent (curative, adjuvant, neoadjuvant, therapeutic, or palliative) will depend on

the tumour type, location, and stage, as well as the general health of the patient.

The radiation �elds may also include the draining lymph nodes if they are clinically

or radiologically involved with tumour, or if there is thought to be a risk of subclin-

ical malignant spread. It is necessary to include a margin of normal tissue around

the tumour to allow for uncertainties in daily set-up and internal tumor motion.

These uncertainties can be caused by internal movement (for example, respiration

and bladder �lling) and accuracy in patient set-up relative to the tumour position.

To spare normal tissues (such as skin or organs which radiation must pass through

in order to treat the tumour), shaped radiation beams are aimed from several angles

of exposure to intersect at the tumour, providing a much larger absorbed dose there

than in the surrounding, healthy tissue.

Shortly after the discovery of the x-ray by German physicist Wilhelm Conrad

Roentgen in 1895, the "powerful rays" were being used to e�ectively treat cancer.

Today, an increasing number of patients have their cancers treated successfully,

with fewer side e�ects and preservation of normal tissue function, using radiation

therapy.

A cancer patient may be treated with radiation alone. Prostate cancer and

larynx cancer are often treated in this manner.

Sometimes radiation therapy is part of a patient's treatment. Patients can be

treated with radiation therapy and chemotherapy before surgery. This may allow a

patient to have less radical surgery than would otherwise be required. Chemotherapy

may be used simultaneously with radiotherapy without surgery to improve the local

response and reduce metastatic disease (9).

There are three fundamental kind of radiation therapy:

� External beam therapy

� Brachytherapy

� Internal therapy, a form of treatment where a source of radiation is put inside

your body.

The treatment planning for external beam therapy consist of:

6

Chapter1. Stereotactic radiosurgery

� de�nition of position and volume of the target and of patience's anatomic

parameter, based on clinical history or imaging technique

� choice of treatment method

� calculation of dose given to the target, and to the surrounding tissues.

This procedure is now computer-assisted by means of appropriate software (8).

1.2 Stereotactic radiosurgery

Stereotactic is a term that means: the involving, being, utilizing, or used in

a surgical technique for precisely directing the tip of a delicate instrument (as a

needle) or beam of radiation in three planes using coordinates provided by medical

imaging in order to reach a speci�c locus in the body.

Stereotactic surgery is a minimally-invasive form of surgical intervention which

makes use of a three-dimensional coordinates system to locate small targets inside

the body and to perform on them some action such as ablation (removal), biopsy,

lesion, injection, stimulation, implantation, radiosurgery (SRS) etc. Stereotactic in

Greek means "solid ordering".

The open stereotactic method provides the basis for radiosurgery and the �rst

stereotactic instrument was designed for use with probes and electrodes. Despite

its name, stereotactic radiosurgery is a non-surgical procedure that delivers a single

high-dose of precisely-targeted radiation using highly focused gamma-ray or x-ray

beams that converge on the speci�c area or areas of the brain where the tumor or

other abnormality resides, minimizing the amount of radiation to health brain tissue.

Although stereotactic radiosurgery is often completed in a one-day session, physi-

cians sometimes recommend multiple treatments, especially for tumors larger than

2,5 cm in diameter. The procedure is usually referred to as fractionated stereotactic

radiosurgery when two to �ve treatments are given and as stereotactic radiotherapy

when more than �ve treatments are given. (11)

During treatment, the patient will lie on a table and the machine may rotate

around the target while it works.

Sometimes, a head frame may be attached to patient's scalp to keep it very still

during therapy. There are di�erent machines used to perform stereotactic radio-

surgery. Some require the use of a frame, and others do not.

7


The entire procedure, including the planning stage, takes about half a day or

less. The time period when patient is receiving the radiation is usually limited.

Some patients receive therapy more than once.

In theory, any organ system inside the body can be subjected to stereotactic

surgery. Di�culties in setting up a reliable frame of reference (such as bone land-

marks which bear a constant spatial relation to soft tissues), however, mean that its

applications have been limited to brain surgery. Plain X-ray images (radiographic

mammography) and computed tomography can be used to guide the procedure.

Conventional external beam radiation therapy � the most common form of radia-

tion therapy � delivers full dose radiation to the tumor and some of the surrounding

brain tissue. For several reasons, the target area for conventional radiation deliber-

ately includes a border (called a �margin�) of normal brain around the tumor. These

reasons include uneven tumor borders, the risk of invisible spread of the tumor into

the surrounding tissue, a larger tumor size, or the presence of multiple tumors. Since

normal brain tissue is included in the full-dose region, conventional radiation is bro-

ken down into small daily doses so the normal brain tissue can tolerate it. As a

result, reaching the desired dose of radiation takes several weeks of daily treatment.

Radiosurgery focuses radiation beams more closely to the tumor than conventional

external beam radiation. This is possible through the use of highly sophisticated

computer-assisted equipment. A head frame or facemask used for this treatment

allows very precise set up, localization and treatment of the tumor. Using advanced

computer planning, radiosurgery minimizes the amount of radiation received by nor-

mal brain tissue and focuses radiation in the area to be treated. Since conventional

radiation therapy covers more normal tissue, it can often be given only once. Ra-

diosurgery, however,may be considered for re-irradiation due to its precision and the

possibility of avoiding previously treated areas.

Radiosurgery is used to treat arteriovenous malformations (AVMs), a tangle of

expanded blood vessels that disrupts normal blood �ow in the brain and sometimes

bleeds. AVMs are the leading cause of stroke in young people. When treated with

radiosurgery, arteriovenous malformations (AVMs) begin to thicken and close o�

slowly, typically over several years.

Stereotactic radiosurgery is an important alternative to invasive surgery, espe-

cially for tumors and blood vessel abnormalities located deep within or close to vital

areas of the brain. Radiosurgery is used to treat many types of brain tumors, either

benign or malignant and primary or metastatic and single or multiple. Sometimes

8


radiosurgery is performed after surgery to treat any residual tumor cells.

Radiosurgery is also a treatment option for other neurological conditions.

Stereotactic radiosurgery works in the same way as other forms of radiation

treatment. It does not actually remove the tumor; rather, it damages the DNA of

tumor cells. As a result, these cells lose their ability to reproduce. Following the

treatment, benign tumors usually shrink over a period of 18 months to two years.

Malignant and metastatic tumors may shrink more rapidly, even within a couple of

months.

There are three basic kinds of stereotactic radiosurgery equipment, each of which

uses di�erent instruments and sources of radiation:

� The Gamma Knife, which uses 192 or 201 beams of highly focused gamma rays

all aiming at the target region. The Gamma Knife is ideal for treating small

to medium size lesions. See the Gamma Knife page for more information.

� Linear accelerator (LINAC) machines, prevalent throughout the world, deliver

high-energy x-rays, also known as photons. The linear accelerator can perform

radiosurgery on larger tumors in a single session or during multiple sessions,

which is called fractionated stereotactic radiotherapy. Multiple manufactur-

ers make this type of machine, which have brand names such as Novalis Tx,

XKnife, and CyberKnife. See the Linear Accelerator page for more informa-

tion.

� Proton beam or heavy-charged-particle radiosurgery. The number of centers

o�ering proton therapy has increased dramatically in the last several years.

(29)

9

Chapter 2

Gamma Knife

Contents

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.2 Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.3 Gamma Knife Perfexion model . . . . . . . . . . . . . . . 13

2.4 Leksell Gamma Plan . . . . . . . . . . . . . . . . . . . . . 16

2.1 Introduction

In the 1950s, Swedish professors Borje Larsson of the Gustaf Werner Institute,

University of Uppsala, and Lars Leksell at the Karolinska Institute in Stockholm,

Sweden, began to investigate combining proton beams with stereotactic (guiding)

devices capable of pinpointing targets within the brain. This approach was eventu-

ally abandoned because it was complex and costly. Instead, in 1967, the researchers

arranged for construction of the �rst Gamma Knife device using cobalt 60 as the

energy source. The prototype unit, used for 12 years in Sweden, was speci�cally

designed for functional neurological surgery, that is, for treatment of patients with

pain, movement disorders, and even certain behavioral disorders that were not re-

sponsive to conventional psychiatric treatment.

Realizing the potential of stereotactic radiosurgery for treating brain tumors,

professor Leksell and his colleagues built a second Gamma Knife in 1975. It was

installed at the Karolinska Institute and became an integral part of the neurosurgical

service there. The third and fourth units, built in the early 1980s, were installed in

10

Chapter2. Physics

Buenos Aires, Argentina, and She�eld, England.

The initial gamma unit design was intended for lesion generation in functional

neurosurgery. Through the development of stereotactic angiography, AVMs became

suitable targets for stereotactic irradiation as did cranial base tumors imaged with

pneumoencephalography or cisternography. In 1980's, an increasing number of pa-

tients had radiosurgery for AVMs, selected benign tumors, and small-volume ma-

lignant tumors. Today over 300000 patients have undergone Gamma Knife surgery

and over 35000 per year receive the treatment.

Currently, the Gamma Knife is used primarily to treat benign brain tumors,

AVMs, acoustic neuromas, pituitary adenomas, craniopharyngiomas, brain metas-

tases, other tumors of the skull base, and pineal region tumors. Most users of

Gamma Knife technology have restricted lesion size to a mean spherical diameter of

35 mm (and usually less). Although larger lesions were treated, to maintain safety

a signi�cant increase in lesion volume must be paralleled by a decrease in delivered

dose. A large decrease in dose for larger lesions leads to a relatively ine�ective total

dose from a radiobiological standpoint, and probably does not improve upon what

might be obtained by standard fractionated techniques. Failure to decrease the dose

for larger volumes can lead to a higher complication rate.

For certain patients with various deep�seated tumors or arterio�venous malfor-

mations, the Gamma Knife may be preferable to conventional surgery. As the unit

can carefully model the radiosurgical doses to lesions of suitable size (preferably less

than three centimeters in diameter), it is used as an alternative approach to stan-

dard microsurgical tools in these patients. Because of the fact the radiation fallo�

is very steep outside the target area, the surrounding brain tissue is spared harmful

after e�ects. Gamma Knife radiosurgery also is safer than many existing procedures

because patients need not undergo risky, open�skull procedures, and adult patients

do not require general anesthesia. Thus, the Gamma Knife is especially useful when

conventional surgical techniques would pose high risk, such as in the presence of

other illnesses or when a patient's age prohibits standard surgery. (28)

2.2 Physics

The basic physics of the Gamma Knife has remained substantially the same since

its conception. The device uses cobalt-60 as a radiation source that decays through

emission to a stable isotope of nickel; 60Ni, with a half life of 5.26 years. As a part

11

Chapter2. Physics

of the decay process, one electron with an energy of up to 315 keV and two gamma

rays with energies of 1.17 MeV and 1.33 MeV are emitted. It is the gamma radiation

that is used to clinical e�ect in the gamma knife and contributes to the naming of

the device.

Figure 2.1: Cobalt-60 decay scheme

The details of the internal design of the gamma knife changes slightly among the

four models currently in use around the world (the U, B, and C models, and the new

Perfexion model). Inside the Gamma Knife unit are an array of 60Co sources (201

sources in the U, B, and C models, 192 in the Perfexion) which are aligned with a

collimation system. The collimation system focuses the individual beams of gamma

radiation to a very precise focus point. While an individual beam has a relatively

low dose rate and causes minimal biological e�ect, the superposition of all beams

at the focus point have a much higher dose rate. The Gamma Knife can therefore

target very precise areas of tissue without causing signi�cant collateral damage to

areas outside of the targeted area.

Modi�cation of the isodose distribution is achieved by using combinations of

isocentres using di�erent collimators, di�erent stereotactic locations, and di�ering

dwell times.

12

Chapter2. Gamma Knife Perfexion model

2.3 Gamma Knife Perfexion model

Figure 2.2: Gamma Knife Perfexion

The Perfexion is the last Gamma Knfe model. The couch serves as the patient

positioning system (PPS), moving the patient to the preselected stereotactic coordi-

nates. The PPS can move between coordinates at speeds of up to 10 mm/s, which is

an order of magnitude faster than the previous Automat Positioning System (APS)

implemented on model C (the �rst model with automatized positioning), which has

a speed of 0.8 mm/s. Docking of the patient to the PPS is achieved by means of an

adaptor that attaches to the standard stereotactic Leksell G-frame with three clips.

The adaptor is then mounted directly to the PPS with a simple locking mechanism.

Angulation of the head in the sagittal plane, the so-called gamma angle, can be at

70, 90, or 110 degrees.

The shielding doors now move horizontally to the left and right, rather than

vertically, before the couch/PPS moves in.

The beam delivery system (radiation unit) has been redesigned with a di�erent

beam geometry. An array of 192 Cobalt-60 sources is arranged in a cone section con-

�guration. This di�ers substantially from the previous hemispherical arrangement.

As a result, the sources have a di�erent source to focus distance, varying from 374

mm to 433 mm. The majority of the sources are closer to the focus/isocenter com-

pared with previous models, yielding a slightly higher dose rate for a given source

13


activity.

Figure 2.3: Gamma Knife Perfexion's collimators

By the inverse square law, beams originating from sources farther from the focal

point contribute a lower dose to the isocentre; thus, more complex modeling is

required to reproduce dose distributions for treatment planning.

The primary and secondary collimators have been replaced by a single, larger, 12

cm thick tungsten collimator array, in which collimators are arranged in a series of

�ve concentric rings, similar to Models B and C. The collimator array is subdivided

into eight independently variable sectors, each containing 24 sources and 72 collima-

tors (24 collimators for each of the three collimator sizes) around its circumference.

The increase in the volume of the radiation cavity to more than three times the

previous volume allows for a greater treatment range in the x and y dimensions of 160

mm and 180 mm, respectively. The z coordinate is limited by the physical distance

from the focal point to the inner surface of the collimator assembly. These ranges

are much greater than the previous models for the automatic positioning system

(APS). This increased distance from focal point to the inner surface, combined

14


with the nearcylindrical shape of the radiation cavity compared with the previously

hemispherical shape of previous Models makes mechanical collisions between the

frame or patient's head and the interior surface of the collimator assembly far less

likely.

The treatment cavity is lined with a 1 mm thick insert made of aluminum,

denoted the collimator cap. This acts as a collision touchguard, causing immediate

retraction of the sources to the home position in response to any pressure being

placed on it, allowing the safe manual removal of the patient. In addition, it protects

the collimator assembly from the ingress of dirt, and it can be easily removed and

cleaned if necessary.

The range of beam diameters has changed from previous models. There are

collimators of three di�erent size 4 mm, 8 mm and 16 mm.

Beam diameters are automatically changed by moving the sources over the se-

lected collimator set, an action performed by servo�controlled motors located in the

sector drive unit at the rear of the radiation unit. Sources have �ve possible posi-

tions: 4 mm, 8 mm, 16 mm, sector o�, and home. Although sector o� and home are

both positions in which the beams are blocked, in sector o�, the sources are closer to

the collimators, taking less than 1 second to reach any of the beam on positions. In

the sector o� position, the sources are located between the 4 and 8 mm collimators,

allowing minimal source travel time to block the beams.

This is used when an individual sector is blocked, when the patient is being

moved between stereotactic coordinates, or when the patient is being transported

into or out of the radiation unit. As a result, the couch no longer needs to move to

the defocus position, as is the case for APS treatments with Model C. This greatly

lowers transit doses to the patient, as beams are only turned on when the patient is

in the correct treatment position. Home is the position occupied by the sources when

the machine is switched o�, idle, or when an emergency stop has been activated.

In this position, the sources are withdrawn further into the radiation unit, several

centimeters away from any of the collimators, giving a lower dose rate outside the

unit.

The sources are licensed and meet the ANSI standard N�542 for medical radio-

therapy sources.

Each of the 192 radioactive sources located in the radiation unit is composed of60Co pellets which are encapsulated in double stainless steel capsules with welded

closures.

15

Chapter2. Leksell Gamma Plan

The total activity at the time of loading Leksell Gamma Knife Perfexion is

approximatively 6000 Ci ≡ 222 TBq (1 Ci = 3,7x1010 Bq). The cobalt sources are

delivered to the site in a specially designed and approved protective container, called

cask. Once loaded into the radiation unit, the sources are maintenance free. They

are handled again only when it is necessary to reload the unit with new sources.

The sources are loaded into the positions within the radiation unit by use of a

loading machine. The loading machine is speci�cally designed to transfer sources

from the shipping cask to the radiation unit with full radiation safety. It is built

almost entirely of lead and weights approximately 12000 kg. Due to the heavy

weight of the equipment, radiation unit, loading machine and cask, the room �oor

must withstand a total weight of 38000 kg. A certain working area is also required.

(24) (30)

2.4 Leksell Gamma Plan

Figure 2.4: Leksell Gamma Plan Perfexion's screenshot

The Leksell GammaPlan PFX (LGP PFX) is a new version (vs 8.2) of the LGP

that runs on a PC platform with a Linux operating system as an SQL client server

16


database. This con�guration accommodates remote planning and unlimited near-

instant access to previous patient treatments. Leksell Gamma Plan PERFEXION

supports DICOM image studies obtained from CT, MRI and PET scans as well

as DICOM angiograms. Using the positioning information provided by the Leksell

Coordinate Frame, these image studies are given precise Leksell location coordinates.

New features in the treatment shot dialogue window include composite shots

and dynamic shaping as well as mouse and keyboard shortcuts, enabling faster,

more user friendly treatment planning.

The most outstanding original feature of the collimator design is the ability to

generate a single isocenter composed of di�erent beam diameters. This composite

shot feature allows each individual isocenter of a treatment plan to have its own

optimized shape, thereby increasing the conformity of the overall treatment plan.

In the treatment of multiple metastases in which a single isocenter may be used

for each tumor, this individual isocenter can now be shaped to the target, without the

time penalty of changing collimator helmets or increasing the number of isocenters

to increase conformity.

Intermediate collimator sizes can be mimicked by alternating sectors with di�er-

ent collimator sizes. For example, a 6 mm collimator can be created by placing 4

and 8 mm collimators in opposing sectors. The ability to change collimators in less

than 1 second removes the previous time penalty of approximately 8 minutes every

time a collimator helmet needed to be changed.

Extreme elongation of isodoses from a single isocenter can be achieved by se-

lective blocking of sectors. This feature allows blocking of more sources than that

previously achievable on Models B and C by beam plugging, allowing greater control

of isodoses and potentially greater dose gradients. Doses to critical structures can

be limited by a process called dynamic shaping. This feature has the same func-

tion as the automatic generation of plug patterns using \textit{shields} in previous

versions of LGP as used for Models U, B, and C. Instead of placing a shield on a

volume at risk, the critical structure is outlined and de�ned as a risk volume. Sec-

tors that contain beams that pass through the risk volume are blocked. The severity

of blocking can be varied between four levels. The sectors are then automatically

blocked during treatment, resulting in no extra set up time, as would be required

with Models U, B, and C. However, beam on times will increase accordingly in order

to compensate for the lower dose rate from the partially blocked collimator array,

just as with traditional plugging. A potential disadvantage with the new system is

17


the inability to block individual beams to protect distant structures, e.g., the lens.

However, the manufacturer's measurements on anthropomorphic phantoms show

lower doses to the eyes as a result of increased shielding and lower transport doses.

For targets closer to the eyes, where larger numbers of beams need to be blocked,

dynamic shaping adapts for protection of the lens. (25)

18

Chapter 3

Isocentre dose measure with

ionization chamber

Contents

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.2 Gamma Knife Perfexion Dosimetry Phantom . . . . . . 20

3.2.1 Dosimetry Phantom With Ionization Chamber . . . . . . 21

3.3 Ionization Chamber . . . . . . . . . . . . . . . . . . . . . . 23

3.3.1 Determination of the absorbed dose . . . . . . . . . . . . 24

3.4 The Bragg-Gray Cavity Theory . . . . . . . . . . . . . . . 25

3.4.1 Bragg-Grey Theory and Ionization Chamber . . . . . . . . 27

3.5 Experimental procedure . . . . . . . . . . . . . . . . . . . 28

3.6 Data discussion . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.6.1 Output factor . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.1 Introduction

The purpose of this chapter is to illustrate the materials, the experimental pro-

cedure and the theory behind the �rst step of this study, the isocentre dose measure

made with a ion chamber.

The experimental part of the study was done in the Niguarda Ca' Granda hos-

pital's Gamma Knife center.

19

Chapter3. Gamma Knife Perfexion Dosimetry Phantom

The �rst test was a punctual dose measure made with a free-air ion chamber,

31016-PTW Freiburg with a small sensible volume, inserted in a phantom made of

Gammex 457 SolidWater (Elekta Dosimetry Phantom) with a diameter of 160 mm,

simulating a simplify standard human head. The ion chamber was at the center of

the phantom placed at the 60Co rays focus point of the GK unit. A dose measure with

all the 16 mm collimators open was made and compared with the Fluka simulation

in the same condition. The error in the �uka simulation was a�ected by the 5 %

error in sources activity and the experimental estimated was 3 %. The agreement

between the two value is inside the margin of error. This experimental measure is

very important, because is the only value we can put in the LGP PFX.

As second step a comparison between the simulated 4 and 8 mm collimators'

punctual dose and the experimental value multiply by the output factor (ratio be-

tween a collimators' calculated dose and the reference 16 mm collimators' dose)

given by Elekta (the constructor company) was made with a good agreement.

3.2 Gamma Knife Perfexion Dosimetry Phantom

Figure 3.1: Leksell Gamma Knife Dosimetry Phantom(courtesy of Elekta)

For the experimental measurements has been used a Gamma Knife Perfexion

dosimetry phantom.

The shape and size of Gamma Knife Perfexion Dosimetry Phantom simulate the

head of an adult human, in fact it is a spherical phantom with a 8 cm ray. Gam-

20


mex 457 Solid Water responds to 60Co radiation �eld as water would do regarding

absorbed dose, and, exactly for this reason this material has been selected to build

the Gamma Knife Perfexion Dosimetry Phantom.

Elekta Dosimetry Phantom allow several con�guration with di�erent components

depending on the application and the measures type. Generally speaking, the main

con�guration of the phantom are two:

� Dosimetry Phantom with ionization chamber, for measuring absorbed dose

rate.

� Dosimetry Phantom �lm stack, for checking precision in dose delivery.

Figure 3.2: Ionization Chamber position inside Gamma Knife Perfexion DosimetryPhantom

3.2.1 Dosimetry Phantom With Ionization Chamber

The con�guration with ionization chamber consists in several components as

show in the following �gure:

21


Figure 3.3: Exploded view of Leksell Gamma Knifer Dosimetry Phantom with ion-ization chamber, mounted on a frame.

The phantom base À is mounted on the frame Á, there are three cylindrical

holes in the ion chamber clamp Â the ion chamber housing Ã can be �tted into the

largest hole. The ion chamber housing is cylindrical and divided into two section by

a mechanical stop. The �lm holder align rods Ä are inserted into the two smaller

holes in the ion chamber clamp. The ion chamber clamp and the upper �lm holder

Å are �xated to the phantom by the front clamp Æ using two screws Ç. (4)

Measures are correctly performed when a small ionization chamber is inserted

in the appropriate phantom set-up, the phantom is attached into the frame adapter

22

Chapter3. Ionization Chamber

and it is moved to the treatment position, corresponding to Leksell coordinates

X = Y = Z = 100mm. In this con�guration the radiological focus point will

coincide with sensitive measuring point of the chamber, �nally the measures of

absorbed dose, according to the dose rate protocol, will be performed with all 192

sources on at 16mm collimators open. (5)

3.3 Ionization Chamber

Experimental measures has been performed with ionization chamber model Pin-

Point PTW 31016 waterproof, this chamber type is simple to use for measurements

in a phantom. The chamber cavity volume is about 0.016 cm3, this size is a com-

promise between the need for su�cient sensitivity and the ability to measure dose

at a point.

Figure 3.4: Ionization chamber PTW31016

PTW-31016 is a cylindrical chamber-type with a 2.9 mm diameter and length.

This chamber shows an identical spatial resolution in all three dimensions due to

the special design, this provides additional precision when scanning beams in the x

and y direction without changing the chamber orientation.

In use, the chamber, must be aligned in such a way that the radiation �uence

is approximately uniform over the cross-section of the cavity-chamber. The cavity

length therefore sets a lower limit on the size of the �eld in which measurements

may be made.

23

Chapter3. Ionization Chamber

The construction of the chamber should be as homogeneous as possible but for

technical reasons the central electrode is likely to be of a material di�erent from

that of the walls. Indeed the choice of materials may play an important role in

ensuring that the energy response of the chamber does not vary considerably. It is

also necessary for the air cavity not to be sealed; it should be designed so that it

will equilibrate rapidly with the ambient temperature and air pressure.

In PTW-31016 the wall material is compound by 0.57mm of PMMA and 0.09mm

of graphite, wall density is about 1.19 g/cm3 for PMMA and 1.84 g/cm3 for graphite,

the central electrode is made of Aluminum 99.98 R.

3.3.1 Determination of the absorbed dose

The absorbed dose to water at the reference point in water-like phantom for a

reference beam of quality Q0 and in the absence of the chamber is given by:

DwQ0 = MQ0ND,w,Q0 (3.1)

whereMQ0 is the reading of the dosimeter under the reference conditions used in the

standards laboratory and ND,w,Q0 is the calibration factor in terms of absorbed dose

to water of the dosimeter obtained from a standards laboratory. In most clinical

situations the measurement conditions do not match the reference conditions used

in the standards laboratory. This may a�ect the response of the dosimeter and

it is then necessary to di�erentiate between the reference conditions used in the

standards laboratory and the clinical measurement conditions.

When a dosimeter is used in a beam of quality Q di�erent from that used in its

calibration, Q0, the absorbed dose to water is given by:

DwQ0 = MQND,w,Q0kQQ0 (3.2)

where the factor kQQ0 corrects for the e�ects of the di�erence between the reference

beam quality Q0 and the actual user quality Q, and the dosimeter reading MQ

has been corrected to the reference values of in�uence quantities, other than beam

quality, for which the calibration factor is valid. The beam quality correction factor

kQQ0 is de�ned as the ratio, at the qualities Q and Q0, of the calibration factors in

24

Chapter3. The Bragg-Gray Cavity Theory

terms of absorbed dose to water of the ionization chamber.

kQQ0 =ND,w,Q

ND,w,Q0

=Dw,Q/MQ

Dw,Q0/MQ0

(3.3)

The most common reference quality Q0 used for the calibration of ionization cham-

bers is 60Co gamma radiation, in which case the symbol kQ is used for the beam

quality correction factor. Ideally, the beam quality correction factor should be mea-

sured directly for each chamber at the same quality as the user beam. However,

this is not achievable in most standards laboratories. Such measurements can be

performed only in laboratories having access to the appropriate beam qualities. For

this reason the technique is at present restricted to a few primary standard dosime-

try laboratories (PSDLs) in the world. The procedure requires the availability of

an energy-independent dosimetry system, such as a calorimeter, operating at these

qualities. A related problem is the di�culty in reproducing in a standards labora-

tory beam qualities identical to those produced by clinical accelerators. When no

experimental data are available, or it is di�cult to measure kQQ0 directly for realistic

clinical beams, in many cases the correction factors can be calculated theoretically.

Where Bragg-Gray theory can be applied, an expression for kQQ0 is:

kQQ0 =(Sw,air)Q

(Sw,air)Q0

(Wair)Q

(Wair)Q0

PQ

PQ0

(3.4)

which depends only on quotients of water to air stopping-power ratios and per-

turbation factors at the beam qualities Q and Q0. The only chamber speci�c factors

involved are the perturbation correction factors PQ and PQ0 . (6)

3.4 The Bragg-Gray Cavity Theory

Fundamental theory of radiation dosimetry and �rst theory developed, Bragg-

Gray cavity theory relates the absorbed dose to the material in a small cavity in

a uniformly irradiated piece of matter, to the absorbed dose to the surrounding

material. Cavity sizes are referred to as small, intermediate or large in comparison

with the ranges of secondary charged particles produced by photons in the cavity

medium. The case where the range of charged particles (electrons and positrons) is

much larger than the cavity dimensions (i.e., the cavity is regarded as small) is of

special interest. Brag-Grey theory is founded on two conditions:

25


1. The cavity must be small when compared with the range of charged particles

incident on it, so that its presence does not perturb the �uence of charged

particles in the medium.

2. The absorbed dose in the cavity is deposited solely by those electrons crossing

the cavity.

If a �uence Φ of identical particles of kinetic energy T passes through an interface

between two di�erent media, g and w, then the absorbed dose on the g-side of

boundaries can be written as:

Dg = Φ

[(dT

ρdx

)c,g

]T

(3.5)

in a similar way, on the w-side:

Dw = Φ

[(dT

ρdx

)c,w

]T

(3.6)

where [(dT/ρdx)c,g]T and [(dT/ρdx)c,w]T are the mass collision stopping-powers of

the two media, evaluated at energy T. The result of �rst condition is that the electron

�uences are almost the same and equal to the equilibrium �uence established in the

surrounding medium, but this condition can only be valid in regions of charged

particle equilibrium or transient charged particle equilibrium.

Second condition implies that:

� Photon interactions in the cavity are negligible and thus ignored.

� All electrons depositing the dose inside the cavity are produced outside the

cavity and completely cross the cavity.

� No secondary electrons are produced inside the cavity and no electrons stop

within the cavity.

Therefore, assuming that the value of Φ is continuous across the interface (i.e. ig-

noring backscattering) the ratio of absorbed doses in the two media adjacent to their

boundary can be written as the ratio of the mass collision stopping-powers of the

two media:

Dw

Dg

=(dT/ρdx)c,w

(dT/ρdx)c,g

. (3.7)

26


Mathematical statement often referred as Bragg-Grey relation.

Where [(dT/ρdx)c,w]/[(dT/ρdx)c,g] is the ratio of the average unrestricted mass

collision stopping powers of the medium and the cavity.

The validity of Bragg-Grey relation, under the term of two Bragg-Grey con-

ditions, holds only for each monoenergetic component of the spectrum for charge

particles crossing the cavity. For a di�erential energy distribution Φt (particles

per cm2MeV ) the appropriate average mass collision stopping-power in the cavity

medium g is:

mS̄g =

∫ Tmax

0

ΦT

( dTρdx

)c,gdT∫ Tmax

0

ΦT dT

=1

Φ

∫ Tmax

0

ΦT

( dTρdx

)c,gdT =

Dg

Φ. (3.8)

And likewise, for a thin layer of wall material w ;

mS̄w =

∫ Tmax

0

ΦT

( dTρdx

)c,wdT∫ Tmax

0

ΦT dT

=1

Φ

∫ Tmax

0

ΦT

( dTρdx

)c,wdT =

Dw

Φ. (3.9)

The Bragg-Gray relation in terms of absorbed dose in the cavity, is then:

Dw

Dg

=mS̄w

mS̄g

≡ mS̄wg . (3.10)

3.4.1 Bragg-Grey Theory and Ionization Chamber

One of the most applications of ion chamber is in the measurement of gamma-ray

exposure. An air-�lled ion chamber is particularly well-studied for this application

because exposure is de�ned in terms of amount of ionization charge create in air.

Under the proper conditions a determination of ionization charge in a air-�lled

ionization chamber can give an accurate measure of exposure, and a measurement

of the ionization current will indicate the exposure rate.

27

Chapter3. Experimental procedure

Dose and exposure are related by the following relation:

Dg =∆Q

∆m

(W̄

e

)g

(3.11)

where, ∆Q is is the charge of the ions of one sign, expressed in Coulomb, collected

from the active volume of the gas during irradiation, ∆m is the mass, expressed in

Kg, of the active volume of the gas, and (W̄/e)g is the average energy deposited in

the gas divided by the charge of one of the ion pairs, per ion pair created.

The concept of active volume just means that volume of the gas for which electric

�eld lines can sweep ions toward the electrodes. There may be guard volumes or in-

active volumes that still logically comprise the cavity volume, but do not participate

in sweeping the ions toward the electrodes.

Therefore, assuming an exposure measurement, we may write for the absorbed

dose in the medium:

Dg =∆Q

∆m

(W̄

e

)g

mS̄wg (3.12)

This equation allows one to calculate absorbed dose in the medium immediately

surrounding a Bragg-Grey cavity, on the basis of the charge produced in the cavity-

gas, provided hat the appropriate values of ∆m, (W̄/e)g, and mS̄wg are known. (7)

3.5 Experimental procedure

In this work the measurements of exposition were performed under Bragg-Gray

conditions, and introducing all corrective factors relative to ionization chamber cal-

culation about pression and room temperature. The Elekta Dosimetry Phantom has

been used with the PinPoint ionization chamber inserted in it and positioned at the

isocentre of the 192 sources of 60Co. The measurement was acquired with all the 16

mm collimators open and the time of exposition was 100.8 s in order to obtain an

absorbed dose of 5 Gy. This experimental measurement is very important, because

it is the only value requested as input by the LGP PFX, when a new radiating device

is installed this measure is asked by the software. The percentage error, for 60Co

photons beam quality Q0, is of 3 % and has been provided by National Laboratories

of Standard Measures.

Because of the di�erences of ambient condition ( temperature and humidity),

28

Chapter3. Data discussion

compared to Standard measures, has been necessary to correct respect work-conditions.

In this study this �rst measure has been compared with the Monte Carlo Fluka

value obtained by the simulation in the same condition. The error in the MC simu-

lation is a�ected by the error in the sources activities, which is the 5 %, compared

to that the statistical error can be neglected.

3.6 Data discussion

Exposition Time(s) Experimental Dose(Gy) MC Dose(Gy) Di�erence

100,8 5±3% 4,95±5% 1%

Table 3.1: Experiment-Simulation comparison

In the table above it is showed the absorbed dose obtained with the ionization

chamber (Experimental Dose) and with the FLUKA simulation (MC Dose). The

last term is the percentage di�erence between the two value:

Difference =Experimental Dose−MC Dose

Experimental Dose· 100 (3.13)

There is a good agreement between the two measure.

3.6.1 Output factor

For all LGK models, the absolute output is calibrated in reference to the dose

rate of the largest open collimator, i.e. 16 mm for the PFX system and 18 mm for

the other LGK models. Such a change largely re�ects the new design of the PFX

system (7).

The output factors (OF) for LGP (version 8.2) are the ratio between the dose

given by a set of collimators and the dose given by the 16 mm set of collimators.

OF =Di

D16

(3.14)

29


where i = 4, 8.

With the experimental value obtained from the ionization chamber, it is possi-

ble to make a �rst test of the LGP . We can multiply the output factors by the

experimental measure made with the ionization chamber (Experimental Dose) and

compare the results with the Fluka MC Dose obtained simulating the 4 and 8 mm

collimators set up.

Collimator(mm)

Outputfactor

Experimentaldose (Gy)

MC Dose(Gy)

Di�erence

8 0,924 4,62 4,54 1,7 %

4 0,805 4,02 3,93 2,2 %

Table 3.2: Output factor comparison

Where Di�erence is the same as in 3.13.

Based in this study, the MC simulation validate the output factor of the LGP.

30

Chapter 4

Relative dose pro�les with

radiochromic EBT

Contents

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 31

4.2 Radiochromic EBT �lms . . . . . . . . . . . . . . . . . . . 32

4.2.1 Composition . . . . . . . . . . . . . . . . . . . . . . . . . 33

4.2.2 Physical and chemical behavior . . . . . . . . . . . . . . . 34

4.3 Leksell Gamma Knife Dosimetry Phantom �lm stack . . 36

4.4 Experimental Procedure . . . . . . . . . . . . . . . . . . . 37

4.4.1 Cutting marking and positioning . . . . . . . . . . . . . . 37

4.4.2 Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

4.4.3 Scan and analysis . . . . . . . . . . . . . . . . . . . . . . . 38

4.4.4 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4.5 Data discussion . . . . . . . . . . . . . . . . . . . . . . . . . 40

4.5.1 Error in the Gafchromic EBT evaluated dose . . . . . . . 40

4.5.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4.1 Introduction

The purpose of this chapter is to illustrate the materials, the experimental pro-

cedure and the theory behind the measures of the relative dose pro�les and their

31

Chapter4. Radiochromic EBT �lms

comparison with the MC simulation's data.

For this measure radiochromic EBT �lms, made by ISP (International Specialty

Products), were used. Gafchromic EBT �lm o�ers many advantages making it

an appropriate instrument for this kind of dosimetry measures. It is near tissue

equivalence, easy to handle in room arti�cial ligth, self-developing and it has high

spatial resolution.

GafCrhomic �lms were cut, marked, placed inside the Elekta Dosimetry Phantom

in �lm stack con�guration by apposite holder and exposed to a rising dose from 0

to 10 Gy in order to obtain a calibration curve to correct the non linear response

to radiation's dose of the �lm. After calibration, three set of measure for coronal

and axial plane were made for all collimators con�guration (16, 8, 4 mm), giving a

5 Gy dose. The �lms were scanned with a scanner (Epson Expression 10000XL) for

a graphical comparison with the MC simulation values. Even in this case a good

agreement between experiment and simulation was achieved.

4.2 Radiochromic EBT �lms

In radiation dosimetry there are several problems associated to the measurement

of isodose curves (lines delimiting body areas receiving the same quantity of radia-

tion in radiotherapy) and depth-dose distributions in high-gradient regions of beams

using conventional measuring system such as ion chambers, semiconductors, ther-

moluminescent detectors and radiographic �lms. These di�culties have resulted in

a search for radiation dosimeter with high spatial resolution which does not require

a special developmental procedure and give permanent absolute values of absorbed

dose with an acceptable accuracy and precision and ease of handling and data anal-

ysis (3).

Some of this feature have been achieved with the introduction of radiochromic

dosimeters. These dosimeters, with very high spatial resolution and relatively low

spectral sensitivity variation are insensitive to visible light, thus o�ering ease of

handling and preparation in room light (4).

Radiochromic EBT �lms has been used for several years as a dosimetry media. It

is near tissue equivalence, and its high spatial resolution makes it ideal for measuring

in a steep dose gradient region. Gafchromic EBT (International Specialty Products,

NJ, USA), radiochromic �lm is one of the newest radiation-induced auto-developing

x-ray analysis �lms available for therapeutic radiation dosimetry in radiotherapy

32


applications.

This �lm does not require any physical, chemical or thermal processing and it's

optical density (OD) is relatively stable within 24 hours, have a dimension of 20,3 x

25,4 cm and they can be use for a dose interval from 0,1 to 8 Gy. GafChromic EBT

tolerate a temperature up to 70oC. Anyway ISP suggest to keep gafchromic �lms at

room temperature (20-25 oC).

Gafchromic ISP-EBT has been developed speci�cally to address the needs of the

medical physicist and dosimetrist working in the radiotherapy environment.

4.2.1 Composition

Figure 4.1: Con�guration of Gafchromic EBT dosimetry �lm

Most radiochromic �lm products have active components, which include a radiation-

sensitive monomer. Upon irradiation, this active monomer polymerizes to form a

polymer colored dye; thus, the automated darkening process upon irradiation with

x-rays takes place. The �nal color, or more correctly the absorption spectrum, is

dependent on the proprietary chemical within the �lm itself. Radiation-induced re-

actions can have an incubation period of at least 1ms and polymerization can proceed

after irradiation has ceased causing a post-exposure density growth which manifests

itself as a signi�cant increase in the optical absorption. This fact is well known

within the medical physics community through the use of various radiochromic �lm

33


products and the general rule of leaving �lms for a period of 24 h before readout is

performed (16).

Gafchromic EBT consist on two active layers 17 µm thick separated by a surface

layer 6 µm thick protected by two polyester layer each one 97 µm thick.

The active layers are water-equivalent material with a Zeff equal to 6,98. The

atomic composition of the GafChromic active layer is H (37,9%), C (42,3%), O

(16,2%), N (1,1%), Li (0,3%) and Cl (0,3%). (27)

4.2.2 Physical and chemical behavior

Gafchromic EBT �lm, has been commercially released which produces a di�erent

absorption spectrum from older radiochromic �lm products. The �lm also produces a

higher dose sensitivity which more closely matches the dose criteria for radiotherapy

applications.

When the active component is exposed to radiation, it reacts to form a blue

colored polymer with absorption maxima at about 636 nm, as seen in Figure 4.2 .

Figure 4.2: Absorption spectra of GAFCHROMIC EBT

Consequently the response of this dosimetry �lm will be enhanced by measure-

ment with red light.

34


The post-exposure density growth of GAFCHROMIC EBT dosimetry �lm has

been investigated. In common with the currently available GAFCHROMIC dosime-

try �lms, the density of the developmental �lm increases with time following expo-

sure. For all GAFCHROMIC dosimetry �lms the density growth is approximately

proportional to log(time after exposure). However, for the EBT �lm the post-

exposure growth is substantially less than for previous GAFCHROMIC �lms. Also

the time period over which signi�cant post-exposure occurs is much less for EBT

�lm. Finally the magnitude and rate of post exposure changes in EBT �lm are

dependent upon the mode of readout (14, 26).

Figure 4.3: Percentage increase in optical density for EBT-�lm post irradiation atvarious dose level

In Cheung et al. (14) work as shown in Figure 4.3 an approximate 7%�9% increase

in coloration occurs up to 24 h post-irradiation. In absolute OD terms the increase

in postirradiation coloration is dependent on the magnitude of the delivered dose.

Approximately 1% of this coloration is from 6 h after irradiation. (3)

35

Chapter4. Leksell Gamma Knife Dosimetry Phantom �lm stack

4.3 Leksell Gamma Knife Dosimetry Phantom �lm

stack

The phantom with �lm stack consist of several components which is shown in

�gure.

Figure 4.4: Exploded view of Elekta Gamma Knife Dosimetry Phantom with vertical�lm stack, mounted on a frame.

The phantom base À is mounted in a frame Á. The ion chamber plug Â is placed

in the bottom hole of the phantom base. The bottom �lm holder Ã, the thin �lm

36

Chapter4. Experimental Procedure

Holder Ä and upper �lm holder Å have engravements on the arched side and are

positioned together by two �lm holder aligne rods Æ. When positioned together the

engravements from a diagonal track across the arched surface. The front clamp Ç

is �xated to the phantom by two screws È. (4)

4.4 Experimental Procedure

This experimental measuring took place at the Niguarda Ca' Granda hospital's

Gamma Knife center. It is a dosimetric measurement of the relative dose pro�le,

transverse dose measurements performed in the coronal plane or axial plane at a

given depth in the phantom, using radiochromic EBT-ISP �lms in the focus point

of the Gamma Knife unit. The position was 100,100,100 in the Gamma Knife

reference system.

4.4.1 Cutting marking and positioning

EBT �lm can be safely handled in the normal indoor lighting before and after

exposure. However is not totally insensitive to light. When the �lm is not being set

up in the phantom, or being exposed to the radiation �elds, or being digitized or

measured, it is recommended to minimize the exposure to light, so the �lms were

kept in their box.

GafChromic EBT were cut with a paper cutter in pieces 6,3 ± 0,3 cm x 4,0 ±0,2 cm. It is essential to scan all pieces in the same direction, for this reason a track

of the orientation relative to the original sheet was kept. The �lms were marked

with a number and Cartesian axis orientation, so it is possible to track the set up

used for the measure. Since the outside surfaces are polyester GafChromic �lm can

be marked using a pen.

Using the Elekta dosimetry phantom in the �lm stack con�guration, radiochromic

EBT �lm was placed in the phantom in the radiative unit for the experimental mea-

sures centered in the focus point. The Gamma Knife PERFEXION reference system

coordinates of the focus point were 100,100,100.

4.4.2 Exposure

For the experimental measurement the EBT �lms were placed in the phantom

inside the GK units at the focus point of the gamma rays of the 60Co sources. Film's

37


exposures in six di�erent con�guration were performed:

1. All 16 mm collimators open and �lm in coronal plane

2. All 16 mm collimators open and �lm in axial plane





A set of three measurement for each con�guration was performed.

4.4.3 Scan and analysis

Figure 4.5: Imagej screenshot

Radiochromic EBT �lms were scanned, after about 24 hours from exposure, by

an Epson Expression 10000XL high quality �at bed scanner. The resolution is 72 dpi

38


and the GafChromics were digitalized with a 16 bit pixel depth as not compressed

TIFF images. The software used for acquisition was �Image Acquisition� by Epson.

In order to optimize the process it is necessary to perform 5 blank scanning to warm

up the Xenon �uorescent lamp. The scanner's borders were not used, because in

the central zone it is possible to obtain a �at response (0,3 % from mean value) and

a mask around the scanning zone was �xed to minimize the di�used light. The gap

between the values read by the scanner is about 1 %.

A scanner's limitation is the saturation level. In fact it is possible to separate

grey scales equivalent to an optical density equal to 2.9-3. For this reason it is

necessary to pay attention to the dose interval during the exposure.

The �rst step is to scan the unexposed piece of �lm, then the exposed pieces

were scanned all with the same orientation. The scanning procedure thus results in

two sets of images: unexposed �lm and irradiated �lms.

After the scanning process the images were analyzed by means of the �ImageJ�

program version 1.40a.

4.4.4 Calibration

Even if this study concerns relative dose pro�les, it is necessary to introduce

a dose calibration for the EBT dosimetry �lm because of the non-linear response

to gamma radiations. Radiochromic �lms were staked in the Elekta Dosimetry

Phantom and irradiated with rising radiation dose from 0 to 10 Gy in order to

obtain a calibration curve necessary to correct the non linear response of the �lms.

The calibration curve is obtained by a �t with an eighth degree polynomial:

y = a+ bx+ cx2 + dx3 + ex4 + fx5 + gx6 + hx7 + ix8 (4.1)

where a = -8,3742x10−6, b = 4,8567x10−5, c = 1,2399x10−8, d = -3,5511x10−12,

e = 4,6299x10−16, f = -3,0954x10−20, g = 1,1261x10−24, h = -2,114x10−29,

i = 1,6127x10−34; r2 = 0.9996.

39


Figure 4.6: Calibration curve

4.5 Data discussion

4.5.1 Error in the Gafchromic EBT evaluated dose

The response curve of a radiochromic �lm is usually expressed by the optical

density (OD) in relation with the given dose, in this case represented by the gray

scale of the scanner.

The optical density is de�ned as:

OD = −log10(I

I0)

where I is the intensity of light at a speci�ed wavelength l that has passed

through a sample (transmitted light intensity) and I0 is the intensity of the light

before it enters the sample (incident light intensity).

The experimental error is a�ected by to many factors like the di�erence of re-

sponse from a �lm to another, the error in the determination of net optical density,

the time for development after exposure.

40


Therefore the experimental error was determined using the Devic et al. method

(14) that was applied by Dr Parini in his work (19) at the Niguarda Ca' Granda

hospital.

The �rst step of this work is to calculate the net optical density and it's standard

deviation in this way:

netOD = log10Iunexp − 65535

Iexp − 65535(4.2)

σnetOD =1

ln10

√σ2

unexp

Iunexp

+σ2

exp

Iexp

(4.3)

where Iunexpand Iexp are respectively the intensity value of the exposed and the

unexposed �lm, σunexpand σexptheir standard deviation.

In order to �nd the most suitable function for the given system, the following

criteria are used:

1. the �t function has to be monotonically increasing;

2. the �t function has to pass through zero;

3. we choose the function that gives the minimum relative uncertainty for the

�tting parameters.

Based on these criteria, the chosen �t function is of the form:

Dfit = b · netOD + c · netODn (4.4)

The term n is introduced to account for the nonlinear dose dependence while

approaching the high dose region close to saturation level for radiochromic EBT-

ISP. The best curve was found with a, b and as �tting parameters and n was varied

from 0,5 to 5 with a step of 0,5.

b σb c σc n r2

558 34 3043 545 3 0,998

Table 4.1: Fitting parameters

41


Finally the error is dependent from the �t parameters and the scanner (1%),

applying the error's propagation theory:

σtot(%) =

√netOD2σ2

b + netOD2nσ2c + (b+ ncnetODn−1)2σ2

netOD + σ2scanner

D· 100

(4.5)

The percent total error depend on dose and it is represented by the curve of �g.

4.7.

Figure 4.7: Uncertainty estimates fof GafChromic EBT �lm (19)

This procedure shows that for a given dose over 0,3 Gy the uncertainty is esti-

mated to be smaller than 8 %.

4.5.2 Results

In the following �gures the graphical comparison between the FLUKA MC sim-

ulation and the radiated radiochromic EBT-ISP �lms data is shown.

The EBT �lms errobars represent the statistical error of the radiochromic �lm,

which in the dose interval of the measurement of this study is under the 8 %, while

the FLUKA errorbars is the statistical error of the simulation.

42


16 mm collimators

The agreement for this collimators set up is inside the margin of error and the

most accurate. In this case the simulation was ran with the largest number of

primaries particles.

The 16mm diagrams are split in two (positive and negative axis), allowing an

easier reading of data.

Coronal plane

20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.8: Film 22

43


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.9: Film 22

20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.10: Film 23

44


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


45


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


Axial plane

20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.14: Gaf 31

46


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.15: Gaf 31

20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.16: Gaf 32

47


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.17: Gaf 32

20

30

40

50

60

70

80

90

100

110

0 2 4 6 8 10 12 14

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.18: Gaf 33

48


20

30

40

50

60

70

80

90

100

110

-14 -12 -10 -8 -6 -4 -2 0

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.19: Gaf 33

8 mm collimators

As shown in the diagrams of �g. 4.20 to 4.25, the statistic for the 8 mm collimator

con�guration is limited. In fact the statistical error is greater than 10 % at the sides

of the diagrams. In this case, it would be usefull to run a simulation with a larger

number of primaries to achieve a more accurate comparison.

49


Coronal plane

20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


50


20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


Axial plane

20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


51


20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


20

30

40

50

60

70

80

90

100

110

-6 -4 -2 0 2 4 6

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


52


4 mm collimators

In the 4 mm collimators con�guration the percentage error is small even at the

sides of the curve, because the statistic is lower than in the centre. The agreement,

especially in the axial plane pro�les, is not so good probably, but the resolution of

radiochromic �lm does not allow to draw de�nitive conclusions.

It is possible to improve the accuracy of the experimental measurements, in

order to reduce the statistical error, repeating the procedure described by Devic et

al. (14). Particularly it is necessary to take �ve times over the same scanning region

as the acquired images with the �lm pieces. This allows one to correct for �defective�

pixels, de�ned as pixels that di�er in intensity from the blank unattenuated signal.

It would be also helpfull to use an higher resolution scanner to improve the resolution

of the images.

Coronal plane

20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


53


20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.27: Gaf 17

20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.28: Gaf 18

54


Axial plane

20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA


55


20

30

40

50

60

70

80

90

100

110

-4 -3 -2 -1 0 1 2 3 4

Dm

ax/D

(%

)

mm

EBT FILMFLUKA

Figure 4.31: �lm 27

56

Conclusion

The FLUKA Monte Carlo simulation of the Gamma Knife Perfexion stereotactic

radiosurgery system is validated by the experiments described in this thesis. A re-

�nement is probably necessary both in the simulation (statistics, geometrical model)

and in the accuracyof the measurements with radiochromic �lm to improve the cor-

respondence between data and simulation. This is a preliminary work for a future

testing of the Leksell Gamma Plan version 8.2 developed by Elekta. As shown in

other publications (1, 17, 18), the LGP algorithm is based in a simpli�ed model

of radiation interaction with matter. In fact the LGP does not consider air cavity,

inhomogeneities and zone of interface. The future prosecution of this study should

include in the analysis of this topic.

The proposed method for the future work is to acquire a CT (Computer Tomog-

raphy) scan of the phantom in order to perform a treatment plan by means of the

LGP PFX software. It is possible to insert in the Elekta Dosimetry Phantom �lm

holders made by di�erent media (aluminum, delrin), or air cavities, and perform CT

scans and new experimental measurements using the modi�ed phantom. It is also

expected to run new simulations with a target reproducing the phantom with the

insertion of inhomogeneous �lm holders. At last, a comparison between the LGP's

treatment planning and the simulated and experimental results will be performed.

By means of the combined results of simulation (computation) and measure-

ments it should be possible to quantify how much the LGP PFX treatment plan

overestimates or underestimates the dose given to the patients. An information of

this kind will be helpfull for the future treatments planning.

57

Bibliography

[1] W. Xiong, D. Huang, L. Lee, J. Feng, K. Morris, E. Calugaru, C. Burman, J

Li, , C-M Ma, �Implementation of Monte Carlo simulation for Gamma Knife�.

Journal of Phisycs: Conference Series 74 (2007).

[2] Fluka User Guide 2008

[3] Azam Niroomand-Rad, C. R. Blackwell, B. M. Cousey, K. P. Gall, J. M. Galvin,

W. L. McLaughlin, A. S. Meigooni, R. Nath, J. E. Rodgers, C. G. Soares,

�Radiochromic �lmdosimetry: Recommendations of AAPM Radiation Therapy

Committee Task Group 55 �. Med. Phys. 25, pp 2093-2115 (1998).

[4] S. Devic, J. Seuntgens, E. Sham, G. Hegyi, E. B. PodGorsak, G. C. Soares, A.

S. Kirov, I. Ali, �Dosimetric property of improved GafChromic �lm for seven

di�erent digitizers�. Med. Phys. Vol. 31 (9), pp. 2392-2401 (2004).

[5] Elekta Instruments, �Leksell Gamma Knife Dosimetry Phantom, Instruction

for use� (2007).

[6] P. Andreo, D.T. Burns, K. Hohlfeld, M.S. Huq, T. Kanai, F. Laitano, V. Smith,

S. Vynkier, �Absorbed Dose Determination in External Beam Therapy: An In-

ternacional Code of Practice for Dosimetry based on Standard of Absorbed Dose

to Water �. IAEA TRS-398 (2004).

[7] H.F. Attix, �Introduction to Radiological Physic and Radiation Dosimetry�. A

Wiley-Interscience Publication, (1986).

[8] L. Ma, P. Kjäll, J. Novtny Jr., H. Nordstrom, J. Johanssön, L. Verhey, �A

simple and e�ective method for validation and measurement of collimator out-

put factors for Leksell Gamma Knife Perfexion�. stacks.iop.org/PMB/54/3897

(2009).

58

BIBLIOGRAPHY BIBLIOGRAPHY

[9] G. Agati, �Introduzione alla �sica radiologica�. Seconda edizione, edizioni libre-

ria cortina (1988).

[10] G. Cittadini, �Diagnostica per immagini e radioterapia�. ECIG (2002).

[11] L. Leksell, �Stereotactic Radiosurgery�. Journal of Neurology, Neurosurgery,

and Psichiatry 46, 797-803 (1983).

[12] J. Novotny Jr, L. Dade Lunsford, M.S. Huq, �Measurement of relative output

factors for the 8 and 4 mm collimators of Leksell Gamma Knife Perfexion by

�lm dosimetry�. Med. Phys. Vol. 36 (5), pp. 1768-1774 (2009).

[13] M. Ruschin, H. Nordström, P. Kjäll, Y.-B. Cho, D. Ja�ray, �Investigation of

intracranial peripheral dose arising from the treatment of large lesions with

Leksell GammaKnife Perfexion�. Med. Phys. Vol. 36 (6), pp. 2069-2073 (2009).

[14] ISP, International Specialty products, �Gafchromic EBT, Self-developing �lm

for radiotherapy dosimetry�. (Agosto 2007).

[15] S. Devic, J. Seuntjens, E. Sham, E.B. Podgorsak, C. Ross Schmidtlein, S. Kirov,

Christopher G. Soares, �Precise radiochromic �lm dosimetry using a �at-bed

document scanner �. Med. Phys. Vol. 32 (7), pp. 2245-2253 (2005).

[16] T. Cheung, M. J. Butson, P.K.N. Yu, �Post-irradiation colouration of

Gafchromic EBT radiochromic �lm�. Phys. Med. Biol. Vol. 50, pp. 281-285

(2005).

[17] V.Moskvin, R. Timmerman, C. DesRosiers, M. Randall, P. DesRosiers, P.

Dittmer, L. Papiez, �Monte Carlo simulation of Leksell Gamma Knife: Source

modelling and calculation in homogeneus media�. Phys. Med. Biol. Vol. 47, pp.

1995-2011 (2002).

[18] V.Moskvin, R. Timmerman, C. DesRosiers, M. Randall, P. DesRosiers, P.

Dittmer, L. Papiez, �Monte Carlo simulation of Leksell Gamma Knife: II. Ef-

fects of heterogeneous versus homogeneous for stereotactic radiosurgery�. Phys.

Med. Biol. vol. 49, pp. 4879-4895 (2004).

[19] F.M. O Al-Dweri, E.L. Rojas, A.M. Lallena, �E�ects of bone- and air-tissue

inhomogeneities on the dose distributions of the Leksell Gamma Knife calculated

with PENELOPE �. Phys. Med. Biol. Vol. 50, pp. 5665-5678 (2005).

59

BIBLIOGRAPHY BIBLIOGRAPHY

[20] V. Parini, �Caratterizzazione di un sistema per trattamenti radioterapici con

modulazione della �uenza del fascio�. Tesi di specializzazione in �sica sanitaria,

Università degli Studi di Milano (2008).

[21] J. Trnka, J. Novotny Jr, J Kluson, �MCNP-based computational model for the

Leksell Gamma Knife�. Med. Phys. 34 (1), pp. 63-75 (2007).

[22] J. Y. C. Cheung, K. N. Yua, C. P. Yu, R. T. K. Ho, �Monte Carlo calculation

of single-beam dose pro�les used in a gamma knife treatment planning system�.

Med. Phys. Vol. 25 (9), pp. 1673-1675 (1998).

[23] Y.C. Cheung, K.N. Yu, R.T.K. Ho, C.P. Yu, �Stereotactic dose planning system

used in Leksell Gamma Knife model-B: EGS4 Monte Carlo versus GafChromic

lms MD-55 �. Applied Radiation and Isotopes Vol. 53, pp. 427-430 (2000).

[24] C. Lindquist, I. Paddick, �The Leksell Gamma Knife Perfexion and comparison

with its predecessors�. ONS-130 Vol. 61 (2007).

[25] Elekta Insruments AB Sweden, �Reference Manual for Leksell GamaPlan�.

[26] M. Yamauchi, T. Tomonaga, O. Nakamura, �Gafchromic �lm dosimetry with a

�atbed color scanner for leksell gamma knife therapy�. Medical Physics Vol. 31

(5), pp. 1243-1248 (2004).

[27] ISP, International Specialty products, www.ispcorp.com.

[28] Elekta instruments, www.elekta.com.

[29] RadiologyInfo, www.radiologyInfo.org.

[30] J. Régis, M. Tamura, C. Guillot, X. Muracciolle, M. Nagaje, D. Porcheron,

�Radiosurgery of the head and neck with the world's �rst fully robotized 192

Cobalt-60 source Leksell Gamma Knife Perfexion in clinical use�. (2006)

60

Aknowledgements

I would like to thank Elekta industries and Dr. Johansson in particular, Prof.

Collice (Head physician of neurosurgery S.C.,Niguarda), Dr. La Camera (Head of

GK centre) and all his sta�, Dr. Torresin (Chief of medical physics C.S.,Niguarda),

Dr. Mainardi (Niguarda), Ing. Agosteo (Politecnico di Milano), Dr. Mairani

(INFN), Dr. Muraro (INFN).

A special thank go to my supervisors Dr. Battistoni and Dr. Brambilla. They

have provided assistance in numerous ways.

I want to express my gratitude to all my family for the support and patience in the

last years.

At last I would like to Fabrizio for sharing worries and di�culties.

61

Documents

Experimental validation of Monte Carlo simulation of Leksell ......of the Gamma Knife Perfexion stereotactic radiosurgery system performed using the Fluka code. The LGP PFX (Leksell