Validation of the Small Animal Biospace Gamma Imager Model Using GATE Monte Carlo Simulations on the Grid Joe Aoun 1,2 , Vincent Breton 2 , Laurent Desbat 1 , Bruno Bzeznik 3 , Mehdi Leabad 4 , Julien Dimastromatteo 5 1 Grenoble Joseph Fourier University, TIMC-IMAG 2 Clermont-Ferrand Blaise Pascal University, LPC 3 Grenoble Joseph Fourier University, CIMENT 4 Biospace LAB, Paris 5 Grenoble Joseph Fourier University, INSERM U877 MICCAI Grid Workshop – New York 2008 1
Introduction Problem: SPECT images have a poor quality Idea: Correction of the attenuation and the scattering from images Key: Modelling the detector and all the physical interactions Solution: Monte Carlo Simulations => accurate model [Buvat I., 2006] Disadvantage: Long computing time Solution: Grid computing [Breton V., 2003] MICCAI Grid Workshop – New York 2008 2
Outline 1) Introduction 2) Tools and Experiments o Monte Carlo Simulations toolkit o CiGri Grid o The small animal gamma camera model o Experimental set-up o Parallelization of the simulations o Validation of the camera model 3) Results 4) Conclusions and Perspectives MICCAI Grid Workshop – New York 2008 3
Monte Carlo Simulation toolkit : GATE G eant4 A A pplication for T T omographic E mission G “ GATE: a simulation toolkit for PET and SPECT ”, S. Jan et al, Phys. Med. Biol., 49 (2004) 4543-4561. based on GEANT4 : a standard simulation package for high energy physics open source and modifiable coded in C++ : more than 200 classes easy to use : simulations are defined and controlled by macros and scripts MICCAI Grid Workshop – New York 2008 4
CIMENT Grid : CiGri Exploits the idle resources of the CIMENT clusters of the University of Grenoble CiGri Server Cluster Batch Scheduler Best-effort: type of jobs that have CiGri User minimum priority and are submitted 1) Submission of a grid ‘bag of tasks’ application only if there is an idle resource. 2) Request of idle resources as a normal user 3) Returns the number of idle resources (ex. 5 nodes) 4) Submission of tasks on the best-effort queue of the batch scheduler. 5) Execution of the tasks as low-priority best-effort (ex. randomly select 5 tasks from the ‘bag’) jobs Local User 6) Local job submission requesting resources 8) Notification of the killing of the tasks (ex. 3 nodes) Use short time 9) Killed tasks re-entered on the ‘bag’ to be 7) Low-priority best-effort jobs re-scheduled for later execution Jobs stopped (3 jobs killed) MICCAI Grid Workshop – New York 2008 5
CiGri infrastructure Resource management system : OAR (http://oar.imag.fr) CiGri software: SQL database interacts with independent modules scheduling jobs submitting jobs cluster synchronizing monitoring jobs collecting results logging errors killing jobs Accessible through a User Interface and monitored through a web portal (https://ciment.imag.fr/cigri) MICCAI Grid Workshop – New York 2008 6
Submission of a job on CiGri JDL USER SSH SSH 1. Macro_i 2. Status_i Cluster 1 User Interface Cluster 2 Output_i Automatically collected Cluster 3 MICCAI Grid Workshop – New York 2008 7
GATE on CiGri Total GATE availability GATE availability (max) (average) Clusters CPUs Clusters CPUs Clusters CPUs Day 11 886 7 430 7 125 Nights and 11 866 7 555 7 215 Weekends MICCAI Grid Workshop – New York 2008 8
Outline 1) Introduction 2) Tools and Experiments o Monte Carlo Simulations toolkit o CiGri Grid o The small animal gamma camera model o Experimental set-up o Parallelization of the simulations o Validation of the camera model 3) Results 4) Conclusions and Perspectives MICCAI Grid Workshop – New York 2008 9
The Biospace small animal Ɣ Imager model light guide Circular field of view crystal D = 10 cm Al protection Continuous NaI(Tl) crystal : D = 12 cm & thickness = 4 mm lead shielding PSPMT = Photomultiplier modelled as a PSPMT capillary ruler 2 mm glass entrance window and a 11 cm nickel backpart collimator LEHR parallel hole collimator with 35 mm thickness holes Septum thickness = 0.2 mm 1.3 mm MICCAI Grid Workshop – New York 2008 10
Experimental set-up: Source in the center of the Field Of View The radioactive background measured first and subtracted from all the other measurements Source placed at different distances from the camera in the air and above a beaker filled with water 10.5 cm Source 99m Tc 16.5 cm 10 cm 11.5 cm 7 cm 4.6 cm Source 99m Tc 2 cm Collimator Crystal NaI(Tl) Crystal NaI(Tl) PSPMT PSPMT MICCAI Grid Workshop – New York 2008 11
Experimental set-up: Source 2 cm off-centered and image of a 4 capillaries phantom Previous measurements were repeated with the source 2 cm off-centered Y Y Concentrations : Capillary • C 1 = 611 µ Ci • C 2 = 220 µ Ci • C 3 = 129 µ Ci • C 4 = 81 µ Ci C 1 Source 1,9 mm centered C 2 C 3 X X 2 cm 2,1 mm C 4 Source off- centered Field Of View 3,15 cm of the camera MICCAI Grid Workshop – New York 2008 12
Parallelization of the simulations Optimization of the camera model: ~ 200 different models were tested 1 configuration 1 big simulation 1 billion emitted events 30 billions random numbers 1 000 000 000 1 million 1 million 1 million 1 million 1 million 1 million 1 million 1 million 1 2 3 … … … … 1 000 1 small simulation 1 million emitted events 30 millions random numbers 10 minutes Local CPU: Pentium IV, 3.2 GHz, 1 Go RAM The Random number streams should be independent [Reuillon R., 2008] MICCAI Grid Workshop – New York 2008 13
Output files of the simulations 1. Retrieved from CiGri 2. Merged into one file on a local CPU 3. Analyzed with the ROOT object oriented data analysis framework (http://root.cern.ch/) MICCAI Grid Workshop – New York 2008 14
Validation of the Ɣ Imager model Comparison of 4 parameters measured experimentally with the corresponding simulated data Energy spectra: events recorded in the whole FOV (40 – 186 keV) Features Features of a of a Sensitivity: Nb of detected events / Nb of emitted events gamma gamma camera camera Spatial Resolution: events recorded in the photopic window 126 – 154 keV Y 6 pixels 6 pixels 6 pixels X FWHM FWHM FWHM FWHM FWHM Image of a capillary phantom: a visual comparison of an inhomogeneous phantom MICCAI Grid Workshop – New York 2008 15
Outline 1) Introduction 2) Tools and Experiments 3) Results o Energy Spectra o Sensitivity o Spatial Resolution o Image of a capillary phantom o CiGri performance 4) Conclusions and Perspectives MICCAI Grid Workshop – New York 2008 16
Energy Spectra MICCAI Grid Workshop – New York 2008 17
Sensitivity in Air Source in Air Experiment 40-186 keV GATE 40-186 keV Experiment 126-154 keV GATE 126-154 keV Experiment 92-125 keV GATE 92-125 keV 7,0E-05 6,0E-05 5,0E-05 Sensitivity (cps/Bq) 4,0E-05 3,0E-05 Relative difference < 5% 2,0E-05 1,0E-05 0,0E+00 2 7 10 16,5 Distance from collimator (cm) MICCAI Grid Workshop – New York 2008 18
Sensitivity with the Beaker Source in Water Experiment 40-186 keV GATE 40-186 keV Experiment 126-154 keV GATE 126-154 keV Experiment 92-125 keV GATE 92-125 keV 6,0E-05 5,0E-05 Relative difference ~ 5% 4,0E-05 Sensitivity (cps/Bq) 3,0E-05 2,0E-05 1,0E-05 0,0E+00 0 4,62 11,55 Water thickness in the beaker (cm) MICCAI Grid Workshop – New York 2008 19
Recommend
More recommend
