IFIMED status and results of a Compton Telescope for hadron therapy
Gabriela Llosá, Enrique Muñoz, John Barrio, José Bernabéu, Ane Etxebeste, Carlos Lacasta, Josep F. Oliver, Pablo G. Ortega*, Carles Solaz.
Instituto de Física Corpuscular – IFIC/IFIMED (CSIC-UV) Valencia, Spain
* and CERN, Geneva, Switzerland
IRIS group: Image Reconstruction, Instrumentation and Simulations in medical applications.
http://ific.uv.es/iris
IFIMED
● Completada la Fase I. Visita de inspección técnica: Abril 2016.
● Instalación asociada orgánicamente al IFIC.
● Dependencias situadas en el parque científico de la UV.
● Pendientes de comenzar la Fase II: acelerador de protones.
IFIMED
● Fase I:
● Obra civil.
● Equipamiento laboratorios de instrumentación para imagen, radiofrecuencia y microPET.
Laboratorio
instrumentación Laboratorio
radiofrecuencia
Micro PET
IFIMED
● MicroPET/CT: SuperArgus de Sedecal.
● SCSIE - Servei Central Suport a Investigacio Experimental.
● Servicios:
● Usuarios finales (investigación biomédica con ratas y ratones, imagenes TAC).
● Usuarios investigación: desarrollo de detectores y algoritmos de imagen.
Sample preparation room
Super Argus PET/CT
IFIMED
● Laboratorio de instrumentación:
● Equipamiento de propósito general (fuentes calibración, cámara climática).
● Investigación en PET, Cámaras Compton y sondas.
X-rays progressively deposit energy as they advance
Protons penetrate the tissue with reduced energy deposition as they advance
Deposit the maximum energy before stopping, defining a peak also known as Bragg peak
Almost no energy deposition beyond the Bragg peak.
HT presents an improved sparing of normal tissue Monitoring of the dose delivery is also more difficult since the beam is fully absorbed
X-rays
protons
IMRT protons carbons
Hadron therapy
● Hadron therapy uses ionizing radiation to treat cancer.
● It is based on the special properties of heavy charged particles such as protons or light ions.
Hadron therapy
● Large benefit over conventional radiation therapies in some cases (ocular tumours, children, organs at risk, radioresistant tumours).
● Higher relative biological effectiveness (RBE) than photons.
● Precise delivery to tumour area => increase of cure rates and reduction of side and long term effects and secondary cancer.
L. Widesott et al. Intensity-
modulated proton therapy versus helical tomotherapy in
nasopharynx cancer: planning comparison and NTCP evaluation.
IJROBP 72(2):589, Oct. 2008
photons Protons
Treatment monitoring
● PROBLEM: the dose administered can not be directly measured (as done in conventional radiotherapy).
● Secondary particles emitted during treatment can be used for monitoring the dose delivery. Essential:
● To verify dose delivery and correct for treatment deviations.
● To reduce safety margins and better exploit hadron therapy.
● Positron Emission Tomography (PET) + MC currently employed.
+ emission is correlated with the dose.
Treatment monitoring
● Dose verification with PET:
Comparison of dose planned and estimated from detected + activity.
● PET Limitations:
● Positron production does not follow irradiation immediately.
● Biological washout- activity carried away by metabolic processes.
● Low amount of + activity induced- low efficiency.
● Difficult online studies – partial ring.
● Photons produce significant background.
Treatment monitoring in hadron therapy
● Prompt gammas emitted from nuclei excited during therapy and can be used for treatment monitoring.
● Emission correlated with dose.
● Emission ~ns after irradiation.
● Emitted in a continuous energy spectrum with energies of MeVs.
Research on collimated and Compton cameras for gamma detection
Energy (MeV)
protons
Telescope PG
Image PG creation map
Compton camera configuration
Problems if the photon energy is unknown or if it can escape (MeV) Scatterer + absorber: 2 interactions.
● Energy determined
● lower efficiency
Multilayer: 3 interactions in 3 detectors (+ correct ordering):
E0=E1+1
2 (E2+
√
E22+4 1−cosE2mecθ2 )cos(θ)=1− E1mec2 E0(E0−E1)
Compton telescope @ IFIC/IFIMED
● Three detector layers:
● Continuous LaBr3 crystals.
– High Compton probability.
– High light yield => good energy and timing resolution.
● SiPM arrays.
● Aim at combination of:
● 2 int events (high efficiency) +
● 3 int events (high resolution).
NO ABSORPTION REQUIRED
Image reconstruction
The cone surface is projected to the reconstruction volume
The intersection of several cone surfaces yields the position of the source
5 events 1000 events
List mode ML-EM
Images shown: 15 iterations, number of events (~5k), Voxel = 1 mm3 Reconstruction of point sources allows performance comparison in different configurations
Three-layer prototype
Detector 1 Detectors 2 and 3
System can measure 2/3 coincidence
events
simultaneously
26.8x27.2x5 mm3 32.4x36.0x5 mm3 32.4x36.0x10 mm3
Detectors
Detector 1 Detectors 2 and 3
36 mm
32.4 mm
27.2 mm
26.8 mm
Crystals
● LaBr3 crystals:
● 27.2 x 26.8 x 5 mm3.
● 32.5 x 35 x 5 mm3.
● 32.5 x 35 x 10mm3.
● Tests with a PMT+ MCA: Energy resolution 3.5% FWHM @ 511 keV
Readout
● VATA64HDR16 ASIC from IDEAS – 64 channels.
● Connected to a DAQ system made at IFIC- Valencia.
● Compact and portable system.
Coincidences board
● Coincidence board based on Xilinx Virtex® FPGA.
● Coincidences between any two or all three planes - still independent image reconstruction.
Prototype
● User-friendly DAQ software also developed at IFIC-Valencia.
Detector characterization
Energy resolution:
7 % FWHM @ 511 keV
Spatial resolution close to 1 mm FWHM
Temperature calibration
1.25 mm FWHM
Na-22 and Y-88
Laboratory tests: radioactive sources
Na-22 Y-88
511 keV
1275 keV
1836 keV 898 keV
● Tests in laboratory performed with two radioactive sources:
● Na-22: 511 and 1275 keV
● Y-88: 898 and 1836 keV
● Imaging capabilities tested in the range [0.5, 1.8] MeV.
Simulations - Measurements
● Experimental measurements reproduced by simulations with GATE 7.0
3.9 mm FWHM Simulation
4.2 mm FWHM Experimental
Very good agreement
Na-22 with two planes
Two planes: different configurations
Measurements with Na-22 selecting 1275 keV peak
4.2 mm FWHM 4.8 mm FWHM 2.6 mm FWHM
4.2 mm FWHM 2.6 mm FWHM
● Increasing distance
between planes improves spatial resolution
● Detection efficiency decreases
● Balance efficiency and resolution
Efficiency = 6.3 * 10-4 Efficiency = 2.1 * 10-3
Two planes: different configurations
Two planes: images with different energies
4.1 mm FWHM 5.7 mm FWHM
4.1 mm FWHM 2.9 mm FWHM
511 keV 1275 keV
1836 keV 898 keV
Na-22
Y-88 Sum Spectrum
Two planes: source at different positions
● Sources placed in different positions, separated 20 mm.
● Points reconstructed separately.
Na-22 selecting 1275 keV peak Y-88 selecting 1836 keV peak
20 mm
Two planes: two sources imaged together
● Sum of energies in both detectors:
898 keV
1836 keV 1275 keV
511 keV
● Selecting 1275 and 1836 keV peaks.
● Sources separated 40 mm.
● Data from both sources reconstructed together.
Sum energy spectrum
Y-88
Na-22 40 mm
Three planes: Na-22
36 mm
40 mm 32 mm
4.1 mm FWHM 5.5 mm FWHM
511 keV 1275 keV
Three planes: Y-88
898 keV
1836 keV
Sum energy spectrum
4.3 mm FWHM
898 keV 1836 keV
2.5 mm FWHM
Tests with protons at KVI-CART
● Tests at KVI-CART, AGOR cyclotron (Groningen).
● Proton beam, 150 MeV, ~ 10 prot/sec.⁸
● Graphite and PMMA targets.
Tests with protons at KVI-CART
● Data with two layers in coincidence.
● PMMA target shifted to simulate Bragg peak variations.
● Shift observed in the Bragg peak position.
P. Solevi et al.
Phys. Med. Biol. vol 61, num
In-Beam Tests with high energy gammas
Tests at HZDR 3MV Tandetron (Dresden).
Daniel Bemmerer1, Fine Fiedler1, Fernando Hueso1, Katja Römer1, Louis Wagner1,2
1 Helmholtz-Zentrum Dresden-Rossendorf (HZDR) , Dresden, Germany.
2 TU Dresden, Dresden, Germany.
Enrique Muñoz, John Barrio, José Bernabéu, Ane Etxebeste, Carlos Lacasta, Josep F. Oliver, Pablo G. Ortega*, Carles Solaz and Gabriela Llosá.
Instituto de Física Corpuscular - IFIC and IFIMED (CSIC-UV) Valencia, Spain
* and CERN, Geneva, Switzerland
Tests with high energy gammas
● Coincidences with 2/3 layers.
Too low rate for three layers.
● Results with layers 1-3
(target at 43 mm, distance between planes 55 mm).
● Telescope placed in three positions wrt target.
Nominal: -10, 0, +10 mm
Reconstructed: -13.7, 0, +10.1 mm
PRELIMINARY RESULTS
Summary and conclusions
●
Three-layer Compton prototype based on continuous LaBr
3crystals and SiPMs constructed at IFIC-Valencia.
●
Studies of detector performance have been carried out with very good agreement with simulations.
●
Successfully reconstructed images from sources in the range [0.5 – 4.4] MeV.
●
First tests in accelerator facilities carried out.
●
Ongoing work is focused on combination of two and
three interaction events and prototype performance
improvement.
Acknowledgements
● This work was supported in part by the European Commission-FP7 through ENVISION project (G.A. num 241851).
● This work was supported in part through the Spanish Ministerio de Economía y competitividad/Plan Nacional de I+D+i (FPA2014-53599-R) and IFIC's Center of Excellence Severo Ochoa (SEV-2014-0398).
● Group members are supported through Ramón y Cajal, Atracció de Talent (UV), Generalitat Valenciana and CPAN contracts.
Thank you!
Related publications:
● P. Solevi et al. Phys. Med. Biol., 2016, volume 61, num 14, 5149-5165
● G. Llosá et al. Front. Oncol., 2016, volume 6: 14
