Advanced

Evaluation of the Catalyst system for patient positioning during breast cancer treatment

Kügele, Malin (2012)
Medical Physics Programme
Abstract (Swedish)
Bröstcancer är den vanligaste formen av cancer hos kvinnor. Fler än 7000 diagnostiseras med denna sjukdom varje år i Sverige. En del av behandlingen mot cancern är strålbehandling. För att strålbehandlingen ska ge goda effekter är det viktigt att det är den sjuka vävnaden som bestrålas, och att den friska vävnaden förskonas från strålningen i den mån det är möjligt. Detta uppnås genom noggranna stråldosplaner individuellt utformade för varje patient. Strålbehandling ges vanligtvis i 16 eller 25 fraktioner, det vill säga att patienten kommer tillbaka dagligen till kliniken 16 eller 25 gånger för behandling. Det är viktigt att patienten ligger i samma position vid varje tillfälle för att stråldosen ska kunna levereras exakt till... (More)
Bröstcancer är den vanligaste formen av cancer hos kvinnor. Fler än 7000 diagnostiseras med denna sjukdom varje år i Sverige. En del av behandlingen mot cancern är strålbehandling. För att strålbehandlingen ska ge goda effekter är det viktigt att det är den sjuka vävnaden som bestrålas, och att den friska vävnaden förskonas från strålningen i den mån det är möjligt. Detta uppnås genom noggranna stråldosplaner individuellt utformade för varje patient. Strålbehandling ges vanligtvis i 16 eller 25 fraktioner, det vill säga att patienten kommer tillbaka dagligen till kliniken 16 eller 25 gånger för behandling. Det är viktigt att patienten ligger i samma position vid varje tillfälle för att stråldosen ska kunna levereras exakt till tumörområdet. I denna studie har ett nytt positioneringssystem utvärderats som heter ”The CatalystTM system” (©2011 C-RAD Positioning AB).

På Skånes Universitetssjukhus i Malmö placeras patienten under behandling liggandes i en ställning på behandlingsbritsen. Armarna placeras i skenor ovanför huvudet och ställningen lutar 7,5⁰ bakom ryggen. Dagens rutiner går till så att under planeringsstadiet i behandlingskedjan har patienten fått tre små tatueringsprickar. Dessa används under positioneringen ihop med laserstrålar som finns i behandlingsrummet. Då prickarna och laserstrålarna sammanfaller ligger patienten i rätt läge. Nästa steg för att försäkra sig om att patienten ligger i rätt position är att ta röntgenbilder på patienten. Man korrigerar patientens position efter resultatet från röntgenbilderna och sedan kan man starta strålbehandlingen. Röntgenbilder tas vanligtvis vid de tre första fraktionerna. Därefter gör man en medelvärdeskorrektion av bordspositionen om det krävs för att få patienten i det korrekta läget. Under behandlingsgången tas även röntgenbild vid en senare fraktion för att försäkra sig om att patienten fortfarande ligger i rätt läge.
The CatalystTM är ett positionerings och övervakningssystem som inte använder sig av röntgen för att kontrollera patientens position utan det här systemet använder sig av optisk scanning av hudytan. Detta har en självklar fördel då det inte bidrar till någon extra stråldos till patienten och kan därmed användas vid varje behandlingstillfälle. Systemet använder sig av en icke-rigid kroppsalgoritm som beräknar hur patienten ska flyttas för att hamna i rätt position. De delar av kroppen som ligger i fel position lyses upp genom att rött eller gult ljus projiceras på patienten, beroende på åt vilket håll som flytten ska ske. Då en arm belyses med rött ljus kan sjuksköterskan enkelt positionera om armen för att få den i rätt position. The CatalystTM har även fördelen att den registrerar eventuella rörelser under bestrålningen. Skulle patienten flytta sig visar systemet detta direkt på en datorskärm så att sjuksköteskorna kan avbryta behandlingen och positionera om patienten.
För att undersöka hur The CatalystTM fungerar i kliniken för att positionera patienter har tre studier utförts. Den första studien var en fantomstudie där fantomet flyttades inom mätvolymen för att undersöka systemets noggrannhet och mätvolymens utsträckning.

Den andra studien utfördes med hjälp att det kliniska CBCT (Cone Beam CT) systemet som tar 3D bilder av skelettet och använder en automatisk benmatchningsfunktion för att ge patientens position. Fantomet i studien hade då inre benstruktur, vilket var mer patientlikt. I studien undersöktes om CBCT systemets automatiska benmatchning och The CatalystTM ytmatchning gav samma positioneringsresultat.
Den tredje studien var en patient studie som innefattade tretton patienter där röntgenbilder och Catalyst bilder togs vid varje behandlingstillfälle för att undersöka hur väl systemen överrensstämmer.

Resultaten från de tre studier som utförts visar att noggrannheten på systemet inte är inom en millimeter vilket är önskvärt. Systemet behöver vidare utveckling för att kunna säkert positionera patienter och The CatalystTM har potential för att lyckas med detta. (Less)
Abstract
Purpose: The CatalystTM system was tested and compared with an X-ray image verification system for patient positioning during breast cancer treatment. Included was to find the optimal reference image and the optimal cropping method for the reference image. Parameters that could lead to an optimization of the treatment routines were also evaluated.

Method and Material: The study was divided into three parts, “Accuracy measurement of the scanning volume”, “The Catalyst system correspondence with CBCT verification images on a phantom” and “The Catalyst system correspondence with planar verification images on patients.”

Accuracy measurements of the scanning volume were performed with a head shaped phantom which had skin equivalent... (More)
Purpose: The CatalystTM system was tested and compared with an X-ray image verification system for patient positioning during breast cancer treatment. Included was to find the optimal reference image and the optimal cropping method for the reference image. Parameters that could lead to an optimization of the treatment routines were also evaluated.

Method and Material: The study was divided into three parts, “Accuracy measurement of the scanning volume”, “The Catalyst system correspondence with CBCT verification images on a phantom” and “The Catalyst system correspondence with planar verification images on patients.”

Accuracy measurements of the scanning volume were performed with a head shaped phantom which had skin equivalent characteristics. The phantom was positioned in a coordinate table and was moved in steps of 2 millimeters in lateral and longitudinal direction in four different planes to investigate the accuracy in the scanning volume.

The CBCT study was performed with a pelvis phantom. The phantom was moved ten times and the deviation from the reference images in the CBCT system and the CatalystTM system were compared.

The Catalyst system correspondence with planar verification images on patients included thirteen patients which at every treatment fraction were positioned with On-Board Imaging (OBI, Varian©), planar verification images (kV). The positioning results were also registered with the Catalyst system. Both systems used the same reference set-up from the CT scan and the positioning results were compared. The optimal cropping method for the reference image was also evaluated.

Results: The results of the study “Accuracy measurement of the scanning volume” showed that the system has a limit at 7.5 cm above the isocenter and that the most accurate results were registered in the plane 5.0 centimeters above the isocenter. The error in the positioning result was 0-4.0 millimeters in the scanning volume. There was no detectable drift in the values in lateral, longitudinal or vertical direction.
The CatalystTM system correspondence with CBCT verification images on a phantom resulted in a high accuracy in positioning in vertical and lateral direction with a correspondence of 0-2.0 millimeters. In the longitudinal direction the results differed between 4.0-6.0 millimeter and which was probably due to a flat structure of the phantom.
The results from the patient positioning study varied depending on the patient. The optimal reference image was determined to be from the CT structure set. An optimal cropping method for the reference image was found and later used for the analysis of the patient positioning.

Conclusions: The CatalystTM system shows accurate positioning result for the phantom studies. The limitation was due to flat or spherical surfaces where the algorithm had difficulties. The flat structure did not provide enough matching information for the algorithm and for the spherical shape the optimization method found a number of solutions. This implies that it is important that the reference image in a patient situation has some structure that the system can use for matching.
In the patient positioning study the reference image for every patient were cropped in an optimal way, mainly to minimize the breathing motion. The results of the study indicated that the system does not correspond well with the planar verification images enough for all patients, possible due to that the verification image system matches due to bony structure and the Catalyst system matches due to the skin surface. (Less)
Please use this url to cite or link to this publication:
author
Kügele, Malin
supervisor
organization
year
type
H2 - Master's Degree (Two Years)
subject
language
English
id
3327125
date added to LUP
2012-12-20 19:08:46
date last changed
2013-09-05 12:22:08
@misc{3327125,
  abstract     = {Purpose: The CatalystTM system was tested and compared with an X-ray image verification system for patient positioning during breast cancer treatment. Included was to find the optimal reference image and the optimal cropping method for the reference image. Parameters that could lead to an optimization of the treatment routines were also evaluated.

Method and Material: The study was divided into three parts, “Accuracy measurement of the scanning volume”, “The Catalyst system correspondence with CBCT verification images on a phantom” and “The Catalyst system correspondence with planar verification images on patients.”

Accuracy measurements of the scanning volume were performed with a head shaped phantom which had skin equivalent characteristics. The phantom was positioned in a coordinate table and was moved in steps of 2 millimeters in lateral and longitudinal direction in four different planes to investigate the accuracy in the scanning volume.

The CBCT study was performed with a pelvis phantom. The phantom was moved ten times and the deviation from the reference images in the CBCT system and the CatalystTM system were compared.

The Catalyst system correspondence with planar verification images on patients included thirteen patients which at every treatment fraction were positioned with On-Board Imaging (OBI, Varian©), planar verification images (kV). The positioning results were also registered with the Catalyst system. Both systems used the same reference set-up from the CT scan and the positioning results were compared. The optimal cropping method for the reference image was also evaluated.

Results: The results of the study “Accuracy measurement of the scanning volume” showed that the system has a limit at 7.5 cm above the isocenter and that the most accurate results were registered in the plane 5.0 centimeters above the isocenter. The error in the positioning result was 0-4.0 millimeters in the scanning volume. There was no detectable drift in the values in lateral, longitudinal or vertical direction.
The CatalystTM system correspondence with CBCT verification images on a phantom resulted in a high accuracy in positioning in vertical and lateral direction with a correspondence of 0-2.0 millimeters. In the longitudinal direction the results differed between 4.0-6.0 millimeter and which was probably due to a flat structure of the phantom.
The results from the patient positioning study varied depending on the patient. The optimal reference image was determined to be from the CT structure set. An optimal cropping method for the reference image was found and later used for the analysis of the patient positioning.

Conclusions: The CatalystTM system shows accurate positioning result for the phantom studies. The limitation was due to flat or spherical surfaces where the algorithm had difficulties. The flat structure did not provide enough matching information for the algorithm and for the spherical shape the optimization method found a number of solutions. This implies that it is important that the reference image in a patient situation has some structure that the system can use for matching.
In the patient positioning study the reference image for every patient were cropped in an optimal way, mainly to minimize the breathing motion. The results of the study indicated that the system does not correspond well with the planar verification images enough for all patients, possible due to that the verification image system matches due to bony structure and the Catalyst system matches due to the skin surface.},
  author       = {Kügele, Malin},
  language     = {eng},
  note         = {Student Paper},
  title        = {Evaluation of the Catalyst system for patient positioning during breast cancer treatment},
  year         = {2012},
}