Simulation and Monte Carlo PDF

Part I Simulation and Monte Carlo Some General Principals and Applications Samira Abbaspour, PhD Department of Medical Physics, Tehran University of Medical Sciences, Tehran, Iran Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION...

Part I Simulation and Monte Carlo Some General Principals and Applications Samira Abbaspour, PhD Department of Medical Physics, Tehran University of Medical Sciences, Tehran, Iran Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … Basics System: The physical process of interest Model: Mathematical representation of the system Models are a fundamental tool of science, engineering, business, etc. Abstraction of reality Models always have limits of credibility Simulation: A type of model where the computer is used to imitate the behavior of the system Monte Carlo simulation: Simulation that makes use of internally generated (pseudo) random numbers Way to Study System System Experiment Experiment actual system model of system Physical Mathematical Model Model Analytical Simulation Model Model Our focus Simulation vs. Analytical Analytical models give exact results, simulation models give approximate results. Simulation is used when an analytical or numerical approach is either impossible or impractical to obtain Analytical models are usually quick to build, but often do not provide a solution with all the actual detail of the problem Simulation models can be built to provide any level of detail, but usually are longer to build and long to run. Some Advantages of Simulation Often the only type of model possible for complex systems Analytical models frequently infeasible Process of building simulation can clarify understanding of real system Sometimes more useful than actual application of final simulation Allows for sensitivity analysis and optimization of real system without need to operate real system Can maintain better control over experimental conditions than real system Time compression/expansion: Can evaluate system on slower or faster time scale than real system Some Disadvantages of Simulation May be very expensive and time consuming to build simulation Easy to misuse simulation by “stretching” it beyond the limits of credibility Problem especially apparent when using commercial simulation packages due to ease of use and lack of familiarity with underlying assumptions and restrictions Slick graphics, animation, tables, etc. may tempt user to assign unwarranted credibility to output Monte Carlo simulation usually requires several (perhaps many) runs at given input values Contrast: analytical solution provides exact values Classification of Simulation Models Static vs. dynamic Static: E.g., Simulation solution to integral  f ( x )dx Dynamic: Systems that evolve over time; simulation of traffic system over morning or evening rush period Deterministic vs. stochastic Deterministic: No randomness; solution of complex differential equation in electrostatic Stochastic (Monte Carlo): Operations of store with randomly modeled radiation transport (SPECT simulation) Continuous vs. discrete Continuous: Differential equations; “smooth” motion of object Discrete: Events occur at discrete times; queuing networks (discrete-event dynamic systems is core subject of books such as Cassandras and Lafortune, 1999, Law and Kelton, 2000, and Rubinstein and Melamed, 1998) Verification and Validation Verification and validation are critical parts of practical implementation Verification pertains to whether software correctly implements the specified model Validation pertains to whether the simulation model (perfectly coded) is acceptable representation  Verification and Validation Validation is the process of comparing two results. In this process, we need to compare the representation of a conceptual model to the real system. If the comparison is true, then it is valid, else invalid. Verification is the process of comparing two or more results to ensure its accuracy. In this process, we have to compare the model’s implementation and its associated data with the developer's conceptual description and specifications. Parallel and Distributed Simulation Simulation may be of little practical value if each run requires days or weeks Practical simulations may easily require processing of 109 to 1012 events, each event requiring many computations Parallel and distributed (PAD) computation based on: Execution of large simulation on multiple processors connected through a network PAD simulation is large activity for researchers and practitioners in parallel computation (e.g., Chap. 12 by Fujimoto in Banks, 1998; Law and Kelton, 2000, pp. 80– 83) Parallel and Distributed Simulation Parallel computation sometimes allows for much faster execution Two general roles for parallelization: Split supporting roles (random number generation, event coordination, statistical analysis, etc.) Decompose model into submodels (e.g., overall network into individual queues) Need to be able to decouple computing tasks Synchronization important—cause must precede effect! Decoupling of airports in interconnected air traffic network difficult; may be inappropriate for parallel processing Certain transaction processing systems (e.g., supermarket checkout, toll booths) easier for parallel processing Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS MCNP4C-BASED X-RAY CT SIMULATOR DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … What is Monte Carlo ? A calculational technique that was pioneered by Stan Ulam and John von Neumann for post-WWII development of thermonuclear weapons Long history of development, particularly n-g One of the original and most important applications of computers Stan Ulam John von Neumann The Monte Carlo Method Statistical probabilities Random numbers (pdfs) Monte Carlo method Computers What is Monte Carlo ? It is a numerical solution to a problem that models objects interacting with other objects or their environment based upon the most essential and simple object-object or object-environment relationships. Intuitive simplicity Solution is determined by random sampling of these relationships until the result converges Repetitive, iterative calculation Examples: Entertainment (games), social science, traffic flow, population growth, game/decision theory, mathematics, finance, genetics, radiation sciences, medical physics, nuclear medicine, radiotherapy, dosimetry... Components of a MC algorithme The primary components of a Monte Carlo simulation method include the following: Probability density functions (pdf's) Random number generator - uniformly distributed Sampling rule Scoring (or tallying) of the quantities of interest Error estimation - an estimate of the statistical error (variance) Variance reduction techniques - reduce the computational time Parallelization and vectorization algorithms Monte Carlo vs Analytic Methods Time to solution Monte Carlo vs Analytic/deterministic approaches Bielajew (2001), Fundamental of MC Method Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS MCNP4C-BASED X-RAY CT SIMULATOR DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … MC in Medical Physics 1600 1400 Number of papers 1200 Radiation protection 1000 Diagnostic radiology 800 600 Radiation therapy 400 Nuclear Medicine 200 0 Other fields …. 1970 1977 1982 1987 1991 1995 1998 2000 Year Number of peer-reviewed publications connected to the applications of Monte Carlo modelling in medical radiation physics published during the last 3 decades Medical Physics 366 Article Physics in Medicine and Biology 371 Article MC Applications in Radiotherapy Linac simulation Gamma Knife simulation Brachytherapy simulation Proton computed tomography simulation Neutron capture therapy simulation MC study of scattered radiation in therapy MC treatment planning Gamma, electron and neutron dosimetry in therapy MC simulation of x-ray spectra in Linac Multi Leaf collimator simulation IMRT and IGRT simulation Portal imaging simulation Linac design optimization Charge particles therapy simulation  MC Applications in Radiotherapy Monte Carlo simulation estimates of neutron doses to critical organs of a patient undergoing 18 MV x-ray LINAC-based radiotherapy , Med. Phys. 32, 3579 (2005) Dose distribution from x-ray microbeam arrays applied to radiation therapy: An EGS4 Monte Carlo study, Med. Phys. 32, 2455 (2005) Monte Carlo simulation of large electron fields at extended distances: Total skin electron treatment optimization, Med. Phys. 32, 2424 (2005) Monte Carlo Investigation of Heterogeneity Effect for Head and Neck IMRT, Med. Phys. 32, 2166 (2005) Monte Carlo Modelling of the Response of NRC's 90Sr/90Y Primary Beta Standard, Med. Phys. 32, 2165 (2005) Error Analysis in Monte Carlo Simulation of Low Energy Brachytherapy Sources, Med. Phys. 32, 2165 (2005) Measurements and Monte Carlo Simulations of Dose Perturbations Due to Metallic Implants in Proton Radiotherapy, Med. Phys. 32, 2164 (2005) Monte Carlo Applications in Conformal, IMRT and 4D Clinical Treatment Planning: Pitfalls and Triumphs, Med. Phys. 32, 2152 (2005) An Efficient Adjoint Monte Carlo Method for Radiation Treatment Planning, Med. Phys. 32, 2071 (2005) Energy and Intensity Modulated Electron Radiation Therapy Using a Monte Carlo Optimization Procedure, Med. Phys. 32, 2070 (2005) MC Applications in Dosimetry Electron dose calculation Gamma and x-ray dose calculation Neutron dose calculation Radiosurgery dose calculation Dose estimation in x-ray imaging Performance prediction of dosimeters Brachytherapy dosimetry Dosimeter design MC Applications in Dosimetry Monte Carlo portal dosimetry, Med. Phys. 32, 3228 (2005) MCPI, An Accelerated Monte Carlo Dose-Calculation Engine for Real-Time Prostate Implant Dosimetry, Med. Phys. 32, 2165 (2005) Implementation of Monte Carlo Dose Verification for Proton Therapy, Med. Phys. 32, 2164 (2005) Monte Carlo Dose Verification for MRI-Based Treatment Planning of Prostate Cancer, Med. Phys. 32, 2147 (2005) An Integrated CT-Based Monte Carlo Dose-Evaluation System for Brachytherapy and Its Application to Permanent Prostate Implant Postprocedure Dosimetric Analysis, Med. Phys. 32, 2068 (2005) Clinical Implementation of Proton Monte Carlo Dose Calculation, Med. Phys. 32, 2028 (2005) Monte Carlo Simulations of the Dosimetric Characteristics of a New Multileaf Collimator, Med. Phys. 32, 2018 (2005) Monte Carlo Calculation of Rectal Dose When Using An Endorectal Balloon During Prostate Radiation Therapy , Med. Phys. 32, 2017 (2005) Verification of Monte Carlo Simulations of Proton Dose Distributions in Biological Media , Med. Phys. 32, 2014 (2005) MC Applications in Radiology MC simulation of x-ray tube Evaluation of Grid performance Detector design Grid design Evaluation of image quality Contribution of scattered radiation in image Dose calculation in radiology imaging X-ray tube ands target/filter optimization Evaluation of digital systems Detection system optimization Detection Quantum Efficiency calculation Calculation of optimal protocol MC Applications in Radiology Scatter/primary in mammography: Monte Carlo validation, Med. Phys. 27, 1818 (2000) Monte Carlo validation in diagnostic radiological imaging, Med. Phys. 27, 1294 (2000) Digital mammography image simulation using Monte Carlo, Med. Phys. 27, 568 (2000) A Monte Carlo study of x-ray fluorescence in x-ray detectors, Med. Phys. 26, 905 (1999) Off-axis x-ray spectra: A comparison of Monte Carlo simulated and computed x-ray spectra with measured spectra, Med. Phys. 26, 303 (1999) Parametrized x-ray absorption in diagnostic radiology from Monte Carlo calculations: Implications for x-ray detector design, Med. Phys. 19, 1467 (1992) Monte Carlo simulation of diagnostic x-ray scatter, Med. Phys. 15, 909 (1988) Monte Carlo simulation of the scattered radiation distribution in diagnostic radiology, Med. Phys. 15, 713 (1988) Monte Carlo studies of x-ray scattering in transmission diagnostic radiology, Med. Phys. 13, 490 (1986) Performance of antiscatter grids in diagnostic radiology: Experimental measurements and Monte Carlo simulation studies, Med. Phys. 12, 449 (1985) MC Applications in CT Performance assessment Optimization of design geometry Scatter characterization Scatter rejection strategies (Septa design) BHE simulation and correction strategies Dose calculation in CT CT design Flat panel CT simulation Optimization of Flat panel CT design Calculation of SPR X-ray Tube design in CT Assessment of correction methods target/filter optimization Calculation of ideal Detector configuration of material Generation raw data for test of reconstruction algorithms MC Applications in CT Monte Carlo Investigation of Scatter Contribution to Kilovoltage Cone-Beam Computed Tomography Images, Med. Phys. 32, 2092 (2005) Development and validation of MCNP4C-based Monte Carlo simulator for fan- and cone-beam x-ray CT, Phys. Med. Biol. 50 No 20 (21 October 2005) 4863-4885 A Monte Carlo based method to estimate radiation dose from multidetector CT (MDCT): cylindrical and anthropomorphic phantoms, Phys. Med. Biol. 50 No 17 (7 September 2005) 3989-4004 Experimental validation of a rapid Monte Carlo based micro-CT simulator, Phys. Med. Biol. 49 No 18 (21 September 2004) 4321-4333 Monte Carlo simulations in CT for the study of the surface air kerma and energy imparted to phantoms of varying size and position, Phys. Med. Biol. 49 No 8 (21 April 2004) 1439- 1454 Reduction of eye lens radiation dose by orbital bismuth shielding in pediatric patients undergoing CT of the head: A Monte Carlo study, Med. Phys. 32, 1024 (2005) Development of a 30-week-pregnant female tomographic model from computed tomography (CT) images for Monte Carlo organ dose calculations, Med. Phys. 31, 2491 (2004) MC Applications in NM/PET Gamma camera simulation Detector modeling and design optimization Scatter characterization Transmission scan modeling Assessment of scatter correction techniques Assessment of attenuation correction techniques Collimator design (Septa in PET) Imaging system design dosimetry in NM Test of reconstruction algorithms Count rate simulation Modeling of electronic performance Detector Block design ( in PET) Motion simulation in NM and PET MC simulation features - PET Hoffman 3D brain phantom Projections Images Ideal Detector «Blurring» Not corrected Corrected MC simulation features - PET Spatial distributions (average over angles): counts counts counts counts bin bin bin bin Measured MC total MC scatter MC unscatter Energy spectra: counts counts counts keV keV keV MC total MC scatter MC unscatter  MC simulation features - SPECT 99mTc 131I Tumour Simulated distribution True distribution 2000 RELATIVE COUNTS non- scatter 1000 total scatter 1st order 3rd 2nd 4th Simulated whole-body images of 131I with the 6th5th activity distribution corresponding to a 131I 0 labelled monoclonal antibody distribution and 40 60 80 100 120 140 160 180 four simulated tumours. ENERGY (keV) Ljungberg et al. (1989) Comp Meth Prog Biomed pp 257-272 MC simulation validation SIMIND - SPECT Eidolon - PET Energy spectra and PSF: measured (Spinks et al 1992) measured (Mazoyer et al 1991) 131I HEGAP measured simulated (Moisan et al 1996) Spectra HEGAP simulated PSF simulated (Zaidi et al 1999) UHEGAP measured UHEGAP simulated 1.0E+04 0.1 Relative Intensity 0.01 Relative Intensity 1.0E+03 0.001 0.0001 1.0E+02 1E-05 100 200 300 400 500 600 700 800 900 100 200 300 400 500 600 700 800 -15 -10 -5 0 5 10 15 Energy (keV) Energy (keV) Distance (cm) 60 Scatter fraction: 50 90 40 Scatter fraction (%) 80 Ljungberg Beck 70 Dresser Ljungberg Scatter fraction (%) 30 measured 60 Beck Dresser simulated (Moisan et al 1996) 50 10 cm depth simulated (Zaidi et al 1999) 40 5 cm depth 20 30 20 10 10 0 0 0.25 1 4 8 16 30 100 150 200 250 300 350 400 Energy w indow w idth (%) Energy (keV) MC simulation validation - PET Brain imaging 2.5 10 4 measured 4 simulated 2 10 4 1.5 10 1 10 4 5000 0 20 40 60 80 100 120 Projection bin Transmission Segmented TX Measured Simulated Spatial distribution Whole-body imaging 5 1.2 10 1 10 5 8 10 4 6 10 4 4 10 4 measured simulated 4 2 10 0 20 40 60 80 100 120 Projection bin Transmission Segmented TX Measured Simulated Spatial distribution MC simulation features - SPECT Zubal torso phantom True activity Ideal image Simulated image Simulated image distribution (in air) [327-400] keV [32-132] keV RSD Heart/Thorax phantom Physical phantom True activity distribution Measured Simulated Dewaraja et al. (2002) Comp Meth Prog Biomed pp 115-124 MC in SPECT collimator’ design Parallel-hole collimator Hoffman brain phantom Parallel-hole collimator Fan-beam collimator FB collimator - optimised for brain Kimiaei et al. (1997) Eur J Nucl Med pp 1398-1404 MC in image reconstruction Jaszczak phantom Reference Analytic (3DRP) Analytic (FORE+2D FBP) Hoffman brain phantom Reference Analytic (3DRP) Iterative (OSEM) Iterative (MRP) MC in image reconstruction - SPECT Activity map Attenuation map Scatter not modeled Approx. scatter PSF MC-based 107 photons Beekman et al. (2002) IEEE Trans Med Imag MC Applications in Nuclear Medicine Relevance of accurate Monte Carlo modeling in nuclear medical imaging, Med. Phys. 26, 574 (1999) A vectorized Monte Carlo code for modeling photon transport in SPECT, Med. Phys. 20, 1121 (1993) Effect of energy resolution on scatter fraction in scintigraphic imaging: Monte Carlo study, Med. Phys. 20, 1107 (1993) A Monte Carlo and physical phantom evaluation of quantitative In-111 SPECT, Phys. Med. Biol. 50 No 17 (7 September 2005) 4169-4185 Fully 3D Monte Carlo reconstruction in SPECT: a feasibility study, Phys. Med. Biol. 50 No 16 (21 August 2005) 3739-3754 Validation of the Monte Carlo simulator GATE for indium-111 imaging, Phys. Med. Biol. 50 No 13 (7 July 2005) 3113-3125 Fast modelling of the collimator–detector response in Monte Carlo simulation of SPECT imaging using the angular response function, Phys. Med. Biol. 50 No 8 (21 April 2005) 1791-1804 Study of the point spread function (PSF) for 123I SPECT imaging using Monte Carlo simulation, Phys. Med. Biol. 49 No 14 (21 July 2004) 3125-3136 MC Applications in PET & PET/CT Validation of GATE Monte Carlo Simulations of the Noise Equivalent Count Rate and Image Quailty for the GE Discovery LS PET Scanner, Med. Phys. 32, 1900 (2005) Validation of a Monte Carlo simulation of the Philips Allegro/GEMINI PET systems using GATE, Phys. Med. Biol. 51 No 4 (21 February 2006) Performance of three-photon PET imaging: Monte Carlo simulations, Phys. Med. Biol. 50 No 23 (7 December 2005) 5679-5695 Monte Carlo simulation and scatter correction of the GE Advance PET scanner with SimSET and Geant4, Phys. Med. Biol. 50 No 20 (21 October 2005) 4823-4840 Unified description and validation of Monte Carlo simulators in PET, Phys. Med. Biol. 50 No 2 (21 January 2005) 329-346 Image quality assessment of LaBr3-based whole-body 3D PET scanners: a Monte Carlo evaluation, Phys. Med. Biol. 49 No 19 (7 October 2004) 4593-4610 Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS MCNP4C-BASED X-RAY CT SIMULATOR DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … GP Monte Carlo Codes Used in Medical Physics MC code General description Language MCNP / General purpose code. Coupled neutrons/photons/electrons transport FORTRAN MCNPX in any material through user generalized geometry. Simulation of imaging systems not specifically included. EGS4 / General purpose code. Coupled photons/electrons transport in any MORTRAN EGSnrc material through user specified geometry. Simulation of imaging systems not specifically included. EGSnrc is an extended and improved version of EGS4. ITS General purpose code. Coupled photons/electrons transport in any FORTRAN material through user specified geometry. Simulation of imaging systems not specifically included. GEANT General purpose code. Coupled photons/electrons transport in any FORTRAN/C++ material through user specified geometry. Simulation of imaging systems not specifically included. PENELOPE General purpose code. Coupled photons/electrons transport in any FORTRAN material through user specified geometry. Simulation of imaging systems not specifically included. FLUKA General purpose code. Coupled photons/electrons transport in any C/C++ material through user specified geometry. Simulation of imaging systems not specifically included. Why MCNP4C? Under developed since 1965, First version in 1977 by Oak Ridge National Laboratory written in FORTRAN language Extensively Tested, Well documented and Widely Used for Simulation of Electron/Photon and Neutron Transport Simulates Accurately all Interactions Very Good Scoring System (Tallying) Powerful Source Distribution Strong Geometry System Rich Collection of variance Reduction Techniques Wide Variety of Adjusting Option (PHYS:P, PHYS:E,….) Support Parallel Processing Monte Carlo Codes Used in X-ray Imaging MC code General description Language ROSI EGS4-based MC package for simulation of x-ray imaging systems. The C++ user interface of ROSI can be used for creation of simulation geometry. Simple predefined shape-based phantoms are available in the code. ViPRIS EGSnrc-based MC package for radiographic image optimization. The VIP- MATLAB Man phantom has been implemented in this computer program as a voxelized phantoms. Unnamed MC software package for x-ray imaging simulation. Photons transport in C any material through shape- and voxel-based phantom. MASTOS MC software package as mammography simulation tool for design Unknown optimization studies. Photons transport in any material through shape- based phantoms. MCMIS MC code for simulation of digital mammography systems. Photons Unknown transport in any material through shape-based phantoms. AMCS Accelerated MC simulator of small-animal micro-CT scanner. Photons Unknown transport in any material through shape- and voxel-based phantoms. CTmod A toolkit for MC simulation of CT scanners. Photons transport inside Unknown simple phantoms for cone-beam single-row detector configuration. Unnamed MCNP4C-based MC simulator for fan- and cone-beam x-ray CT. The GUI MATLAB includes several shape- and voxel-based phantoms and x-ray spectra database. Monte Carlo Codes Used in NM Imaging MC code General description Language SIMSET MC simulator for SPECT and PET imaging systems. Photons transport in C material through voxel-based phantoms. SIMIND MC simulator for SPECT imaging systems. Photons transport in any FORTRAN material through voxel-based phantoms. SIMSPECT MCNP-based MC simulator for SPECT imaging systems in shape- and FORTRAN/C voxel-based phantoms. MCMATV MC simulator for SPECT imaging systems. Photons transport in material FORTRAN through voxel-based phantoms. PETSIM MC simulator for PET imaging systems. Photons transport in material FORTRAN through shape-based phantoms. EIDOLON MC simulator for PET imaging systems. Photons transport in any material Objective-C through shape- and voxel-based phantoms. PET-EGS EGS4-based MC simulator for PET imaging systems in shape- and voxel- FORTRAN/C based phantoms. Unnamed MC simulator for SPECT imaging systems. Photons transport in material FORTRAN/C through shape-based phantoms. GATE GEANT4-based simulator for SPECT and PET imaging systems in shape- C++ and voxel-based phantoms. PET-SORTEO MC simulator for PET imaging systems. Photons transport in any material C through shape- and voxel-based phantoms. Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … Development of Anthropomorphic Phantoms Conceptually, the purpose of a physical or computerized phantom for MC modeling is to represent an organ or body region of interest, to allow modeling the chemical composition of the attenuating medium which absorbs and scatters the x-ray and gamma radiation in a manner similar to biological tissues. In other terms, a phantom is a mathematical model designed to represent an organ or tissue of the body, an organ system, or the whole-body. Stylized mathematical phantoms Tomographic voxel-based phantom Anthropomorphic Phantoms Shape-based phantoms Voxel-based phantoms ORNL MIRD Shepp-Logan VIP-Man Chest Image (MCNP) (MCNP) (Scan2MCNP) A wide panorama of phantoms Mathematical phantoms Voxel phantoms The size and shape of the body Based on digital images and its organs are described recorded by mathematical expressions from scanning of real persons representing combinations and by computer tomography (CT) intersections of planes, or circular and elliptical cylinders magnetic resonance imaging spheres, cones, tori, … (MRI). NRPB Mathematical Phantom Gibbs Phantom (National Radiological Protection Board) (1984) MIRD5 Phantom NORMAN Phantom (Medical Internal Radiation Dose (MRI data of a volunteer) Committee, pamphlet no 5) ORNL Phantom Zubal Phantom (Oak Ridge National Laboratory) (from CT and MRI data)  Mathematical phantoms Spherical Cow is an excellent guide to the very important art of quickly finding approximate answers to problems. In the late 1950’s, the human body was modeled by a sphere, with lots of little spherical organs. Makes it easier to calculate numbers but not too realistic ….. ICRP Publication 2 (1959) Report of Committee II on Permissible Dose for Internal Radiation  Anthropomoprhic phantoms In the late 1950’s, the human body was modeled by a sphere, with lots of little spherical organs Not too realistic ….. Need for improved models … Discussions between MIRD and ICRP: Common effort in developing phantom series 2 main classes of anthropomorphic phantoms: Mathematical (analytical) phantoms: regular/irregular shapes Digital (voxel-based) phantoms: CT/MRI ICRP Publication 2 (1959) “Report of Committee II on Permissible Dose for Internal Radiation” Synder et al. (1969 & 1978) “MIRD Pamphlet No. 5 – MIRD Heterogeneous phantom” Analytical Model First attempts to model the shape of a human being and its internal organs in order to calculate absorbed radiation doses were made by Snyder et al. (1969) and Koblinger (1972). These were based on the anthropomorphic MIRD-type phantom, which was originally developed for the dosimetry of internal radionuclide sources. The Medical Internal Radiation Dose Committee creates MIRD5 that has been the basis for various derivations, like the ORNL Mathematical Phantom Series. The anatomies of new-born, and children of age 1 year, 5 year, 10 year, 15 year and adult male and female had been modelled at the Oak Ridge National Laboratory by Cristy and Eckerman (1987). W. S. Snyder, M. R. Ford, G. G. Warner, H. L. Fisher jr MIRD Pamphlet # 5 Revised: “Estimates of absorbed fraction for monoenergetic photon sources uniformly distributed in various organs of a heterogeneous phantom”, J Nucl Med Suppl 3, 1969. K. F. Eckerman, M. Cristy, J. C. Ryman “The ORNL Mathematical Phantom Series”, http://homer.ornl.gov/vlab/VLabPhan.html  Modelling with geometries Skeletal system An example of how MIRD5 describes brain Exterior of the phantom Brain  “Final” result The sizes and the positions of geometrical representations of body parts are taken from the document: ORNL Mathematical Phantom Series From ORNL Series: model of trunk as elliptical cylinder This document allows to create different phantoms choosing its age and sex. Back view Front view Female ORNL Anthropomorphic Phantom Geant 4 Female ORNL Anthropomorphic Phantom Skull Thyroid Spine Lungs Esophagus Breasts Arm Bones Spleen Heart Pancreas Liver Stomach Kidneys Upper Large Intestine Pelvis Ovaries Uterus Lower Large Intestine Urinary Bladder Leg Bones Not visible: Brain (in the skull) Female ORNL Anthropomorphic Phantom Particle: gamma Energy: 100. MeV no. Particle: 20 Beam Direction: along Z axis Visualization system: OpenGL Output of run 1 TrackID: 2 -> LegBonesORNLVolume -> Energy deposit: 3.4576726 MeV TrackID: 2 -> BodyVolume -> Energy deposit: 5.1323606 MeV TrackID: 2 -> BodyVolume -> Energy deposit: 5.1711365 MeV TrackID: 2 -> BodyVolume -> Energy deposit: 701.06096 keV TrackID: 16 -> BodyVolume -> Energy deposit: 807.59738 keV TrackID: 23 -> BodyVolume -> Energy deposit: 1.497982 keV TrackID: 22 -> BodyVolume -> Energy deposit: 2.3391091 keV TrackID: 21 -> BodyVolume -> Energy deposit: 220.25512 eV TrackID: 20 -> BodyVolume -> Energy deposit: 10.468365 keV TrackID: 19 -> LegBonesORNLVolume -> Energy deposit: 16.651015 keV TrackID: 24 -> BodyVolume -> Energy deposit: 4.0614721 keV Female ORNL Anthropomorphic Phantom Particle: gamma Energy: 100. MeV no. Particle: 20 Beam Direction: along X axis Visualization system: OpenGL Output of run 2 TrackID: 19 -> ArmBonesORNLVolume -> Energy deposit: 2.148 keV TrackID: 34 -> ArmBonesORNLVolume -> Energy deposit: 70.168538 keV TrackID: 33 -> BodyVolume -> Energy deposit: 19.493675 keV TrackID: 32 -> BodyVolume -> Energy deposit: 42.318809 keV TrackID: 31 -> SpleenORNLVolume -> Energy deposit: 119.96815 keV TrackID: 30 -> SpleenORNLVolume -> Energy deposit: 3.9230135 MeV TrackID: 29 -> PancreasORNLVolume -> Energy deposit: 1.0284905 MeV TrackID: 18 -> BodyVolume -> Energy deposit: 543.1 eV TrackID: 37 -> BodyVolume -> Energy deposit: 33.841459 keV TrackID: 36 -> LiverORNLVolume -> Energy deposit: 986.16275 eV TrackID: 35 -> LiverORNLVolume -> Energy deposit: 1.9230902 keV Organ Specifications Mathematical phantoms Cerebrospinal Fluid Spinal Skeleton Cranium Eyes Facial region Teeth Spinal Mandible CSF Thyroid Spinal Column MIRD 5 Revised ORNL Phantom Series New MIRD Head and Brain Models Snyder et al (1978) Cristy (1980) Bouchet et al (1999) MIRD No. 15 Mathematical phantoms The MIRD5 phantom has been the basis for various derivations representing infants and children of various ages (Cristy 1980), gender-specific adult phantoms (Kramer et al 1982) and a pregnant female adult phantom (Stabin et al 1995). Body height and weight as well as the organ masses of these MIRD- type phantoms are in accordance with the Reference Man data (ICRP 1975). Figure 1 shows two versions of derived MIRD5 phantoms known under the names ADAM and EVA phantom (Kramer et al 1982). Mathematical phantoms Cerebrospinal Fluid Spinal Skeleton Cranium Eyes Facial region Teeth Spinal Mandible CSF Thyroid Spinal Column New MIRD Head and Brain Models Bouchet et al (1999) MIRD No. 15 Reference man (ICRP) Segars et al (2001) IEEE Trans Nucl Sci  Mathematical phantoms Left lung Right Lateral View Beating heart Anterior View Dynamic Non-uniform Rational B Spline-based torso phantoms (NURBS) Segars WPL et al (2001) IEEE Trans Nucl Sci 48: 89-97  Complex Shape Based Phantoms Equations are Available at: http://www.imp.uni-erlangen.de/phantoms Abdomen Hip Jaw Resolution Head Thorax A wide panorama of phantoms Mathematical phantoms Voxel phantoms The size and shape of the body Based on digital images and its organs are described recorded by mathematical expressions from scanning of real persons representing combinations and by computer tomography (CT) intersections of planes, or circular and elliptical cylinders magnetic resonance imaging spheres, cones, tori, … (MRI). NRPB Mathematical Phantom Gibbs Phantom (National Radiological Protection Board) (1984) MIRD5 Phantom NORMAN Phantom (Medical Internal Radiation Dose (MRI data of a volunteer) Committee, pamphlet no 5) ORNL Phantom Zubal Phantom (Oak Ridge National Laboratory) (from CT and MRI data) Interest on Anthropomorphic Phantoms 2005, April: Monte Carlo Topical Meeting, Tennessee In the session about “Tomographic Models for Radiation Protection Dosimetry”, many talks about anthropomorphic phantom (mainly voxel-based models) have been presented: - GSF Male And Female Adult Voxel Models Representing ICRP Reference Man By Keith Eckerman - Effective Dose Ratios For The Tomographic Max And Fax Phantoms By Richard Kramer - Reference Korean Human Models: Past, Present and Future By Choonsik Lee - The UF Family of Paediatric Tomographic Models By Wesley Bolch and Choonik Lee - Development And Anatomical Details Of Japanese Adult Male/ Female Voxel Models By T. Nagaoka - Dose Calculation Using Japanese Voxel Phantoms For Diverse Exposures By Kimiaki Saito - Stylized Versus Tomographic Models: An Experience On Anatomical Modelling At RPI By X. George Xu - Use Of MCNP With Voxel-Based Image Data For Internal Dosimetry Applications By Michael Stabin - Application Of Voxel Phantoms For Internal Dosimetry At IRSN Using A Dedicated Computational Tool By Isabelle Aubineay-Laniece - The Use Of Voxel-Based Human Phantoms In FLUKA By Larry Pinsky - The Future Of Tomographic Modelling In Radiation Protection And Medicine (Panel discussion) IMAGE CODING PIXEL Picture Element A digital image is defined as a set of “Pixels” arranged in a “Matrix” of Row “Rows” and “Columns”. At each “Pixel” is attached a number representing the gray level of this point Column An image can be stored in a computer in the form of a series of numbers, one for each Pixel of the Matrix MATRIX SIZE Matrix 300 x 300 Matrix 75 x 75 Matrix 38 x 38 pixel size = 0.5 mm pixel size = 2 mm pixel size = 4 mm The matrix size must be adapted to the spatial resolution and the FOV Usual values CT/MR/Cardio: 512 x 512 RAD/R&F/Vascular: 1k x 1k Chest/Mammo: 2k x 2k  Voxel… What is Voxel? When you are reading images which voxel would you prefer? 10x10x10 0.75x0.17x0.17 0.8x0.35x0.35 A B 5x5x5 1.25x1.25x1.25 F 0.625x0.625x0.625 E C D  Athropommorphic phantoms Voxel-based Zubal phantoms MRI head Phantom CT torso + head Phantom Modified - arms folded Original Modified - arms down Zubal IG and Harrell CR (1994) Med Phys 21: 299-302 Dawson T W et al (1997) Bioelectromagnetics 18: 478–90 Sjogreen et al (2001) J Nucl Med 42: 1563-1570  VIP-man phantoms Xu et al. Health Phys (2000) GSF phantom series GOLEM : voxel model of an adult male CHILD: voxel model of a seven year old girl (selected organs) MEETMAN voxel phantom http://www-ibt.etec.uni-karlsruhe.de/forschung/meetman/meetman_en.htm FAX: a Female Adult voXel phantom FAX: a Female Adult voXel phantom for Monte Carlo calculation in radiation protection dosimetry The main set of data used for the construction of the FAXphantom consisted of 151 consecutive CT images of a 37 year old female patient. The patient’s height was 165 cm, and her weight was 63.4 kg. The images covered the trunk, the neck and the lower part of the head including the mandible with the lower teeth. The pixel size was 0.073 cm × 0.073 cm, and the distance between two consecutive images was 0.5 cm. MAX: a male adult voxel phantom The MAX (Male Adult voXel) phantom has been developed from existing segmented images of a male adult body, in order to achieve a representation as close as possible to the anatomical properties of the reference adult male specified by the ICRP. Scan2MCNP (CT Scan to MCNP Conversion) http://www.whiterockscience.com/scan2mcnp 4D NCAT and XCAT Phantom Related Community http://www.virtualphantoms.org/index.html Analytical Versus Voxelized Phantom Voxelized vs. Mathematical phantoms Mathematical Voxel-based phantoms phantoms Simple geometric descriptions +++ - Realistic patient-specific anatomical descriptions +/- +++ Memory requirements +++ --- Discretization errors +++ -- Modeling dynamic processes ++ -- On-line MC simulation of patient image data --- +++  Physical phantoms Geometrical and brain phantoms Elliptical ECT phantom (DSC) Hoffman 3D brain phantom (DSC) Striatal phantom (RSD) Heart/Thorax phantoms Heart/Thorax phantom (RSD) Static/dynamic anthropomorphic thorax/heart phantom Outline SIMULATION: BASIC AND PRINCIPALES MONTE CARLO SIMULATION APPLICATION OF MC SIMULATION IN MEDICAL PHYSICS MONTE CARLO CODES USED IN MEDICAL PHYSICS DEVELOPMENT OF ANTHROPOMORPHIC PHANTOMS FUTURE APPLICATIONS OF MONTE CARLO … Future Applications of Monte Carlo Future developments in system design, together with faster computer systems and parallel processing will improve: Image quality Temporal resolution Patient throughput and quantification It is expected that Monte Carlo calculations will enter the clinical and scientific arena and will become a method of choice to developed and implement: Patient specific dosimetry Image correction techniques Optimize instrumentation Optimize clinical protocols Zaidi (1999) Med Phys 26: 574-608 Recommended Readings REVIEW PAPERS Andreo P "Mote Carlo techniques in medical radiation physics" Phys Med Biol (1991) 36: 861-920 Zaidi H "Relevance of accurate Monte Carlo modeling in nuclear medical imaging" Med Phys (1999) 26: 574-608 Buvat I and Castiglioni I "Monte Carlo simulations in SPET and PET" Q J Nucl Med (2002) 46: 48-61 BOOK / BOOK CHAPTERS "Monte Carlo calculations in nuclear medicine: diagnostic applications" Eds. Ljungberg M, Stand S-E and King MA (1998) Institute of Physics Publishing, Bristol. ISBN "Therapeutic application of Monte Carlo calculations in nuclear medicine" Eds. Zaidi H and Sgouros G (2002) Institute of Physics Publishing, Bristol. ISBN 0 7503 8168 Thanks For Your Attention