Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Absorbed dose distributions in the vicinity of high-density materials in head and neck radiotherapy A quantitative comparison between measurements, Monte Carlo simulations and treatment planning system

Bjelkengren, Ulf (2007)
Medical Physics Programme
Abstract (Swedish)
Varje år diagnostiseras i Sverige ungefär 1100 personer med cancer i huvudhalsregionen. Typiska diagnoser är tungcancer, läppcancer och cancer i struphuvudet.

I och runt om munhålan finns det stora skillnader i densitet mellan till exempel vävnad och tänder och vävnad och tandlagningar. I detta arbete så har effekterna av ett antal olika tandlagningsmaterial på dosberäkningarna undersökts. För att göra detta så har mätningar jämförts med dosberäkningar från ett kommersiellt dosplaneringssystem samt med avancerade datorsimuleringar.

För att göra mätningarna så byggdes en vattenbalja i vilken tandmaterial av komposit, porslin, titan och guld placerades. Denna vattenbalja bestrålades sedan med en strålbehandlingsutrustning som används... (More)
Varje år diagnostiseras i Sverige ungefär 1100 personer med cancer i huvudhalsregionen. Typiska diagnoser är tungcancer, läppcancer och cancer i struphuvudet.

I och runt om munhålan finns det stora skillnader i densitet mellan till exempel vävnad och tänder och vävnad och tandlagningar. I detta arbete så har effekterna av ett antal olika tandlagningsmaterial på dosberäkningarna undersökts. För att göra detta så har mätningar jämförts med dosberäkningar från ett kommersiellt dosplaneringssystem samt med avancerade datorsimuleringar.

För att göra mätningarna så byggdes en vattenbalja i vilken tandmaterial av komposit, porslin, titan och guld placerades. Denna vattenbalja bestrålades sedan med en strålbehandlingsutrustning som används för att behandla cancer till vardags. Mätningar av absorberad dos framför och bakom tandmaterialet gjordes. För att jämföra dessa mätningar med dosplaneringssystemet så gjordes en datortomografiundersökning av vattenbaljan med de olika tandmaterialen i. Denna datortomografiundersökning användes sedan som grund för dosberäkningen i dosplaneringssytemet på samma sätt som man skulle gjort för en patient. Som referens till dessa två användes resultat från en avancerad datorsimulering. Tyvärr är dessa datorsimuleringar så tidskrävande att de inte går att använda till dosplanering för annat än forskningssyfte och för enkla geometrier som till exempel vattenbaljor.

Resultaten visade stora skillnader mellan mätningarna och dosplaneringssystemet. Datorsimuleringarna och mätningarna visade god överensstämmelse. Skillnaderna mellan mätningarna och dosplaneringssystemet var störst precis framför tandmaterialet samt precis efter. Framför allt för guld så var resultaten från dosplaneringssytemet grovt underskattat precis framför tandmaterialet. Underskattningar av dosen efter tandlagningarna för övriga material kunde härledas till datortomografiundersökningen och begränsningar i dosplaneringssystemet.

Slutsatsen av detta arbete är att det finns begränsningar i dagens dosplaneringssystem som användaren måste känna till. Ett exempel av dessa är att man bör säkerställa att ens datortomografiundersökning är optimal. Detta kan göras med metoder från detta arbete. Även begränsningar i dosplaneringssystemet för hur detta hanterar material av höga densiteter bör uppmärksammas för användaren. För guldtandlagningen och den stora skillnad i dos framför lagningen finns det dock ett ännu enklare sätt att minimera dosen. Detta görs förslagsvis genom att placera en distans av plast, några millimeter tjock mellan tanden med lagning och omgivande vävnad i munnen. (Less)
Abstract
"Introduction: Around 1.000 people are diagnosed with head and neck cancer in Sweden every year. Radiotherapy is often used to treat these tumors. When irradiating patients in the head and neck area there are some problems. One of these problems is the large difference in density between the oral tissue and dental materials. Another problem is the image artifacts due to high-density materials in the CT-images used for radiotherapy treatment planning. These effects must be studied to assure that an optimal treatment is given

Materials and methods: Dental materials were placed at 37 mm depth in a custom built water filled phantom and irradiated using photon beam energy of 6 MV. TLD and a compact ion chamber were used for obtaining depth... (More)
"Introduction: Around 1.000 people are diagnosed with head and neck cancer in Sweden every year. Radiotherapy is often used to treat these tumors. When irradiating patients in the head and neck area there are some problems. One of these problems is the large difference in density between the oral tissue and dental materials. Another problem is the image artifacts due to high-density materials in the CT-images used for radiotherapy treatment planning. These effects must be studied to assure that an optimal treatment is given

Materials and methods: Dental materials were placed at 37 mm depth in a custom built water filled phantom and irradiated using photon beam energy of 6 MV. TLD and a compact ion chamber were used for obtaining depth dose curves before and after the high-density dental material at clinical relevant depths. Monte Carlo simulations were used for reference. Monte Carlo simulations were obtained using the EGSnrc user code DOSXYZnrc. The dental materials studied were composite, porcelain, titanium and a 75 % gold alloy. The sizes of the dental material were 1 cm3 except for the gold alloy that was 0.26 cm3. A computer generated phantom equal to the real phantom was created in the treatment planning system (TPS). The TPS used in this study was the Oncentra Master Plan v1.5 (Nucletron B.V. Veenendaal, The Netherlands). Two dose calculation algorithms have been evaluated, the pencil beam and the collapsed cone. Two different CT scanners have been evaluated in this study.

Results: The compact ion chamber gave a much better agreement with Monte Carlo simulations than the TLD. An overestimation of 23 % in absorbed dose was found just before the gold alloy compared to the result calculated by the TPS. Also an underestimation of dose was found at 5 cm depth for the gold cavity. Differences of up to 8.8 % in absorbed dose were found between the two calculation algorithms. Differences of up to 700 HU was found between the CT-scanned dental materials and the Hounsfield scale implemented in the TPS to calculate electron density. 3.0 % difference in absorbed dose was found at 5 cm depth for the gold alloy between the two different CT-scanners.

Conclusions: The 23 % difference in dose before the gold cavity due to backscattered contaminating electrons cannot be neglected. The use of bite-blocks can minimize the dose to oral tissue due to these electrons. The collapsed cone is the preferable dose calculation algorithm since it calculates dose to the actual medium. CT-scanners must be evaluated before commissioning to find the most suitable filters and reconstruction algorithms to minimize image artifacts. Limitations in the TPS such as assigning electron density and maximum density handled leads to underestimations and overestimations of absorbed dose after high-density cavities" (Less)
Please use this url to cite or link to this publication:
author
Bjelkengren, Ulf
supervisor
organization
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Strålterapi
language
English
id
2157049
date added to LUP
2011-09-14 12:03:41
date last changed
2013-09-05 10:23:50
@misc{2157049,
  abstract     = {{"Introduction: Around 1.000 people are diagnosed with head and neck cancer in Sweden every year. Radiotherapy is often used to treat these tumors. When irradiating patients in the head and neck area there are some problems. One of these problems is the large difference in density between the oral tissue and dental materials. Another problem is the image artifacts due to high-density materials in the CT-images used for radiotherapy treatment planning. These effects must be studied to assure that an optimal treatment is given

Materials and methods: Dental materials were placed at 37 mm depth in a custom built water filled phantom and irradiated using photon beam energy of 6 MV. TLD and a compact ion chamber were used for obtaining depth dose curves before and after the high-density dental material at clinical relevant depths. Monte Carlo simulations were used for reference. Monte Carlo simulations were obtained using the EGSnrc user code DOSXYZnrc. The dental materials studied were composite, porcelain, titanium and a 75 % gold alloy. The sizes of the dental material were 1 cm3 except for the gold alloy that was 0.26 cm3. A computer generated phantom equal to the real phantom was created in the treatment planning system (TPS). The TPS used in this study was the Oncentra Master Plan v1.5 (Nucletron B.V. Veenendaal, The Netherlands). Two dose calculation algorithms have been evaluated, the pencil beam and the collapsed cone. Two different CT scanners have been evaluated in this study.

Results: The compact ion chamber gave a much better agreement with Monte Carlo simulations than the TLD. An overestimation of 23 % in absorbed dose was found just before the gold alloy compared to the result calculated by the TPS. Also an underestimation of dose was found at 5 cm depth for the gold cavity. Differences of up to 8.8 % in absorbed dose were found between the two calculation algorithms. Differences of up to 700 HU was found between the CT-scanned dental materials and the Hounsfield scale implemented in the TPS to calculate electron density. 3.0 % difference in absorbed dose was found at 5 cm depth for the gold alloy between the two different CT-scanners.

Conclusions: The 23 % difference in dose before the gold cavity due to backscattered contaminating electrons cannot be neglected. The use of bite-blocks can minimize the dose to oral tissue due to these electrons. The collapsed cone is the preferable dose calculation algorithm since it calculates dose to the actual medium. CT-scanners must be evaluated before commissioning to find the most suitable filters and reconstruction algorithms to minimize image artifacts. Limitations in the TPS such as assigning electron density and maximum density handled leads to underestimations and overestimations of absorbed dose after high-density cavities"}},
  author       = {{Bjelkengren, Ulf}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Absorbed dose distributions in the vicinity of high-density materials in head and neck radiotherapy A quantitative comparison between measurements, Monte Carlo simulations and treatment planning system}},
  year         = {{2007}},
}