Development of Simulation Framework for Advanced Radiation Therapy

1
Development of Simulation Framework for Advanced
Radiation Therapy

• Takashi Sasaki
• KEK
• (This project is supported by JST CREST)

2
Background

• This is the joint project of Genat4 developers
  and also users
• Japanese Geant4 developers are having an interest
  on medical application of Geant4 since we started
  to work
• We proposed JST for the project to develop
  software framework for simulation in radiation
  therapy based on Geant4 together with medical
  users and also space users
• JST agreed to provide us the funding for this
  research activity from Oct. 2003 to Sep. 2008
• JST Japan Science and Technology Agency
• A governmental funding agency to promote science
  and technology

3
project objectives

• Provide a software framework and necessary
  components for simulation based on Geant4 in
  radiation therapy, especially focused on heavy
  ion therapy
• dose calculation
• Provide a stand alone complete application for
  usage at HIMAC as an example
• with analysis and visualization
• We are NOT intending to develop a treatment
  planning system itself

4
knowledge DB
Analysis Tools
GRID Engine
visualization
framework for medical application
UserActions
Geant4
modeler
PhysicsLists
EGS4
JQMD
Other
5
Member List

• Geant4 developers
• Y.Watase, K. Amako, K. Murakami (KEK),
  H.Yoshida(Naruto), S. Tanaka(RITS), H.
  Kurashige(Kobe) and myself
• Medical
• N. Kanamatsu, M. Komori, N. Matsufuji and T.
  Kanai(NIRS)
• Ken Yusa (Medical School of Gunma Univ.)
• K. Maruyama, T. Hasegawa and H.
  Muraishi(Kitasato)
• Space
• Y. Ozaki, Y. Maeda and K. Nakazawa(JAXA)
• HEP
• S. Kawabata, K. Ukai (KEK) and T. Aso (Toyama)
• Computer Science
• Y. Iida, S. Suzuki, S. Yashiro (KEK) and A.
  Kimura (RITS)

6
Groups

• Software design and integration
• Mainly G4 developers
• Physics validation for heavy ion interaction
• NIRS, Kitasato and Gunma
• Modeling and visualization
• Naruto, RITS and Toyama
• GRID
• KEK Computing Research Center

7
Current activities

• Overall design
• Use-case driven approach
• Physics validation
• comparison of Geant4 with HIMAC (carbon) and
  HIBMC (proton) data
• HIBMC Hyogo Ion Beam Medical Center, another
  heavy ion therapy facility in Japan
• Survey and research on modeling and
  visualization
• a treatment planning system can be used for
  visualizing the result, but we need an impressive
  eye catchy picture to demonstrate our results to
  public and funding agencies
• GRID
• Survey on LCG2 and SRB

8
Plan

• JFY2004 (April 2004March 2005)
• hire 4 researchers
• one at NIRS, one at RITS and two at KEK
• Complete 1st version of the design document
• Comparison between Geant4 and data
• Introducing a small PC farm
• 50PCs or around

9
Items to be discussed

• Geant4 in medical application
• Physics lists and parameters setting
• a set of PhysicsLists based on use-cases
• e.g. proton in water phantom, carbon in emulsion
  and etc
• tips and pit holes
• Geant4 physics
• heavy ion physics
• validation
• Geant4 performance
• benchmarking
• push Geant4 collaboration to improve calculation
  speed of EM processes together

10

• CAD interface in Geant4
• We need it or not??
• If necessary, what we can do? Just wait?
• Fast simulation
• Ray tracing and parameterization
• Current EM physics is too slow to use for
  treatment planning
• Event parallelism may not be sufficient to earn
  computation time
• Use-cases
• Medical user forum
• as like space
• Any other?

11
Acknowledgement

• We thank Maria Grazia Pia and INFN stuffs for
  arranging this meeting
• We hope this opportunity will be the first step
  toward collaboration among us