Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
ONLINE RANGE MONITORING IN
HADRONTHERAPY WITH THE INSIDE
PET SCANNER PhD: Veronica Ferrero
Tutor: Dr. Piergiorgio Cerello XXXI PhD cycle
HADRONTHERAPY
2
ATREP, Trento Cyclotron
220 MeV/u protons Since 2014
CATANA, Catania Cyclotron
62 MeV/u protons Since 2002
“Radiation therapy is the medical use of ionizing radiation to treat cancer. When the irradiating beams are made of charged particles (protons and other ions, such as carbon), radiation therapy is called Hadrontherapy.”
The European Network for LIGht ion Hadron Therapy
~80 operating centers in the world 41 new centers in the next 2 years
CNAO - TERA Foundation, Pavia Synchrotron
60-250 MeV/u protons 120-400 MeV/u carbon ions
Protons since 2011 Carbon ions since 2012
[email protected], INFN and University of Torino
18th March 2017
TREATMENT ACCURACY ASSESSMENT
[email protected], INFN and University of Torino
PHOTONS PROTONSBETHE-BLOCH EQUATION
Mirabell R et al., Potential reduction of the incidence of radiation-induced second cancers by using proton beams in the treatment of pediatric tumor, Int. Jour. Rad. Onc. Phys. 2002, 54 (3) 824
PHOTONS PROTONS
HEART HEART
TREATMENT ACCURACY ASSESSMENT
[email protected], INFN and University of Torino
BETHE-BLOCH EQUATION
tumor underdosage
healthy tissues overdosage
dose uncertainties: organ motion, misalignment, treatment plan
approximations, etc…
Mirabell R et al., Potential reduction of the incidence of radiation-induced second cancers by using proton beams in the treatment of pediatric tumor, Int. Jour. Rad. Onc. Phys. 2002, 54 (3) 824
PHOTONS PROTONS
HEART HEART
PARTICLE RANGE VERIFICATION
5
Fiedler F., et al. Online irradiation control by means of PET. Ion Beam Therapy Fundamentals, Technology, Clinical Applications. Berlin:
Springer-Verlag (2012) p. 527-43
E Palomares et al. Study of the reliability of the cross sections used to model the production of PET isotopes with proton beams. Phys. Med. Biol. 2011; 56:2687-98
Zhu X, Fakhri GE. Theranostics. 2013;3(10):731-740.
off-room PET in-beam PETin-room PET
[email protected], INFN and University of Torino
PARTICLE RANGE VERIFICATION
6
Fiedler F., et al. Online irradiation control by means of PET. Ion Beam Therapy Fundamentals, Technology, Clinical Applications. Berlin:
Springer-Verlag (2012) p. 527-43
E Palomares et al. Study of the reliability of the cross sections used to model the production of PET isotopes with proton beams. Phys. Med. Biol. 2011; 56:2687-98
Zhu X, Fakhri GE. Theranostics. 2013;3(10):731-740.
off-room PET in-beam PETin-room PET
[email protected], INFN and University of Torino
loss of signal due to short half lives, higher washout
7
‣ Bimodal system (tracker under characterization, PET installed at CNAO)
‣ Integrated in the treatment room
‣ Provides a feedback during the treatment
in-beam PETtracker
[email protected], INFN and University of Torino
National Center of Oncological Hadrontherapy (CNAO), Pavia, Italy
THE INSIDE IN-BEAM PET
8
PET heads assembled and tested
January 2016, INFN Torino
Completed PET detector (running)
[email protected], INFN and University of Torino
March 2016, CNAO, Pavia
November 2017, INFN Torino
CHARACTERIZATION TESTS
9
INSIDE @ CNAO
[email protected], INFN and University of Torino
FDG, 68Ge rods PMMA, anthropomorphic phantom
JINST
2017 JINST 12 C03051
P�������� �� IOP P��������� ��� S���� M�������
R�������: December 12, 2016A�������: February 17, 2017
P��������: March 13, 2017
���� T������ S������ �� I��������� P������� ��� R�������� D���������–� O������ ����S����, I����
The INSIDE project: in-beam PET scanner system features
and characterization
V. Ferrero on behalf of the INSIDE collaboration
Department of Physics, University of Torino,Via Pietro Giuria 1, Torino, ItalyIstituto Nazionale di Fisica Nucleare INFN, Sezione di Torino,Via Pietro Giuria 1, Torino, Italy
E-mail: [email protected]
A�������: The INSIDE collaboration has recently completed the construction of an in-beam PETscanner, now under commissioning at the Italian National Center of Oncologic Hadrontherapysynchrotron facility in Pavia. In-beam PET is one of the options for real-time monitoring of theBragg peak range in hadrontherapy sessions, crucial to treatment quality assessments. The systemcharacterization is ongoing and first measurements with clinical beams showed the capability of theINSIDE PET to operate during irradiation delivery and to reconstruct the beam-induced activitymap in real-time. The acquired data were compared to the simulations, with very promising results.
K�������: Instrumentation for hadron therapy; Control and monitor systems online; Imagereconstruction in medical imaging
c� 2017 IOP Publishing Ltd and Sissa Medialab srl doi:10.1088/1748-0221/12/03/C03051
2017 JINST 12 C03051
P�������� �� IOP P��������� ��� S���� M�������
R�������: December 12, 2016A�������: February 17, 2017
P��������: March 13, 2017
���� T������ S������ �� I��������� P������� ��� R�������� D���������–� O������ ����S����, I����
The INSIDE project: in-beam PET scanner system features
and characterization
V. Ferrero on behalf of the INSIDE collaboration
Department of Physics, University of Torino,Via Pietro Giuria 1, Torino, ItalyIstituto Nazionale di Fisica Nucleare INFN, Sezione di Torino,Via Pietro Giuria 1, Torino, Italy
E-mail: [email protected]
A�������: The INSIDE collaboration has recently completed the construction of an in-beam PETscanner, now under commissioning at the Italian National Center of Oncologic Hadrontherapysynchrotron facility in Pavia. In-beam PET is one of the options for real-time monitoring of theBragg peak range in hadrontherapy sessions, crucial to treatment quality assessments. The systemcharacterization is ongoing and first measurements with clinical beams showed the capability of theINSIDE PET to operate during irradiation delivery and to reconstruct the beam-induced activitymap in real-time. The acquired data were compared to the simulations, with very promising results.
K�������: Instrumentation for hadron therapy; Control and monitor systems online; Imagereconstruction in medical imaging
c� 2017 IOP Publishing Ltd and Sissa Medialab srl doi:10.1088/1748-0221/12/03/C03051
JINST
JMI
1
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distributionof this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
The INSIDE project: on-line monitoring and
simulation validation with the in-beam PET scanner
V Ferrero1,2 on behalf of the INSIDE collaboration1 Department of Physics, University of Torino, Via Pietro Giuria 1, Torino, Italy2 Istituto Nazionale di Fisica Nucleare INFN, Sezione di Torino, Via Pietro Giuria 1, Torino,Italy
E-mail: [email protected]
Abstract. The quality assurance of particle therapy treatment is a fundamental issue thatcan be addressed by developing reliable monitoring techniques and indicators of the treatmentplan accuracy. Monitoring using Position Emission Tomography (PET) systems is the onlyin-vivo non invasive technique employed clinically and has been carried out in particle therapysince 1997. However, the PET monitoring of β+ emitter isotopes is typically done after thetreatment, resulting in a large fraction of lost data because of the isotopes rapid physical decay.The INSIDE collaboration has recently installed an in-beam PET scanner at the Italian NationalCenter of Oncologic Hadrontherapy in Pavia, Italy. Here, there is an ongoing project in orderto start testing the method on patients. This work focuses on the online performances of thescanner with clinical beams.
1. IntroductionParticle range verification is one of the key issues for treatment quality assessment inhadrontherapy, and requires the development of reliable monitoring techniques and indicators ofthe treatment plan accuracy. Charged particles release all of their energy inside the body, witha sharp ionization peak at the end of their range called the Bragg Peak (BP), so the only non-invasive way to verify the treatment accuracy is to collect the secondary radiation generatedby the beam-tissue interactions. Positron Emission Tomography (PET) has been employedclinically since 1997 [1], and offers the possibility to directly monitor the spatial distributionof positron (β+) emitter isotopes created during irradiation. Specifically, the emitted positronannihilates with an electron in the tissue, producing two back-to-back 511 keV photons, which inturn can interact and release part (i.e., Compton scattering) or all (photoelectric effect) of theirenergy in the detection crystals. Two detectors triggering in time coincidence allow the definitionof the line of response associated to a certain annihilation event. To correctly identify the linesof response the scattered events need to be discarded, so only the 511 keV events (correlated tothe photoelectric effect) are selected. The reconstructed isotopes activity distribution inside thepatient thus gives an insight about the delivered treatment.
Defining the particle range from the activity measurement is not straightforward, becauseit has a non-trivial dependency on the ion species, the energy, and the tissue composition.However, from the activity distribution it is possible to infer the BP position since its profilealong the beam path has a characteristic fall-off a few millimeters before the BP itself [2, 3, 4].
1
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distributionof this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
The INSIDE project: on-line monitoring and
simulation validation with the in-beam PET scanner
V Ferrero1,2 on behalf of the INSIDE collaboration1 Department of Physics, University of Torino, Via Pietro Giuria 1, Torino, Italy2 Istituto Nazionale di Fisica Nucleare INFN, Sezione di Torino, Via Pietro Giuria 1, Torino,Italy
E-mail: [email protected]
Abstract. The quality assurance of particle therapy treatment is a fundamental issue thatcan be addressed by developing reliable monitoring techniques and indicators of the treatmentplan accuracy. Monitoring using Position Emission Tomography (PET) systems is the onlyin-vivo non invasive technique employed clinically and has been carried out in particle therapysince 1997. However, the PET monitoring of β+ emitter isotopes is typically done after thetreatment, resulting in a large fraction of lost data because of the isotopes rapid physical decay.The INSIDE collaboration has recently installed an in-beam PET scanner at the Italian NationalCenter of Oncologic Hadrontherapy in Pavia, Italy. Here, there is an ongoing project in orderto start testing the method on patients. This work focuses on the online performances of thescanner with clinical beams.
1. IntroductionParticle range verification is one of the key issues for treatment quality assessment inhadrontherapy, and requires the development of reliable monitoring techniques and indicators ofthe treatment plan accuracy. Charged particles release all of their energy inside the body, witha sharp ionization peak at the end of their range called the Bragg Peak (BP), so the only non-invasive way to verify the treatment accuracy is to collect the secondary radiation generatedby the beam-tissue interactions. Positron Emission Tomography (PET) has been employedclinically since 1997 [1], and offers the possibility to directly monitor the spatial distributionof positron (β+) emitter isotopes created during irradiation. Specifically, the emitted positronannihilates with an electron in the tissue, producing two back-to-back 511 keV photons, which inturn can interact and release part (i.e., Compton scattering) or all (photoelectric effect) of theirenergy in the detection crystals. Two detectors triggering in time coincidence allow the definitionof the line of response associated to a certain annihilation event. To correctly identify the linesof response the scattered events need to be discarded, so only the 511 keV events (correlated tothe photoelectric effect) are selected. The reconstructed isotopes activity distribution inside thepatient thus gives an insight about the delivered treatment.
Defining the particle range from the activity measurement is not straightforward, becauseit has a non-trivial dependency on the ion species, the energy, and the tissue composition.However, from the activity distribution it is possible to infer the BP position since its profilealong the beam path has a characteristic fall-off a few millimeters before the BP itself [2, 3, 4].
JPCS
CHARACTERIZATION TESTS
10
INSIDE @ CNAO
[email protected], INFN and University of Torino
FDG, 68Ge rods PMMA, anthropomorphic phantom
JINST
2017 JINST 12 C03051
P�������� �� IOP P��������� ��� S���� M�������
R�������: December 12, 2016A�������: February 17, 2017
P��������: March 13, 2017
���� T������ S������ �� I��������� P������� ��� R�������� D���������–� O������ ����S����, I����
The INSIDE project: in-beam PET scanner system features
and characterization
V. Ferrero on behalf of the INSIDE collaboration
Department of Physics, University of Torino,Via Pietro Giuria 1, Torino, ItalyIstituto Nazionale di Fisica Nucleare INFN, Sezione di Torino,Via Pietro Giuria 1, Torino, Italy
E-mail: [email protected]
A�������: The INSIDE collaboration has recently completed the construction of an in-beam PETscanner, now under commissioning at the Italian National Center of Oncologic Hadrontherapysynchrotron facility in Pavia. In-beam PET is one of the options for real-time monitoring of theBragg peak range in hadrontherapy sessions, crucial to treatment quality assessments. The systemcharacterization is ongoing and first measurements with clinical beams showed the capability of theINSIDE PET to operate during irradiation delivery and to reconstruct the beam-induced activitymap in real-time. The acquired data were compared to the simulations, with very promising results.
K�������: Instrumentation for hadron therapy; Control and monitor systems online; Imagereconstruction in medical imaging
c� 2017 IOP Publishing Ltd and Sissa Medialab srl doi:10.1088/1748-0221/12/03/C03051
2017 JINST 12 C03051
P�������� �� IOP P��������� ��� S���� M�������
R�������: December 12, 2016A�������: February 17, 2017
P��������: March 13, 2017
���� T������ S������ �� I��������� P������� ��� R�������� D���������–� O������ ����S����, I����
The INSIDE project: in-beam PET scanner system features
and characterization
V. Ferrero on behalf of the INSIDE collaboration
Department of Physics, University of Torino,Via Pietro Giuria 1, Torino, ItalyIstituto Nazionale di Fisica Nucleare INFN, Sezione di Torino,Via Pietro Giuria 1, Torino, Italy
E-mail: [email protected]
A�������: The INSIDE collaboration has recently completed the construction of an in-beam PETscanner, now under commissioning at the Italian National Center of Oncologic Hadrontherapysynchrotron facility in Pavia. In-beam PET is one of the options for real-time monitoring of theBragg peak range in hadrontherapy sessions, crucial to treatment quality assessments. The systemcharacterization is ongoing and first measurements with clinical beams showed the capability of theINSIDE PET to operate during irradiation delivery and to reconstruct the beam-induced activitymap in real-time. The acquired data were compared to the simulations, with very promising results.
K�������: Instrumentation for hadron therapy; Control and monitor systems online; Imagereconstruction in medical imaging
c� 2017 IOP Publishing Ltd and Sissa Medialab srl doi:10.1088/1748-0221/12/03/C03051
JINST
JMI
1
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distributionof this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
The INSIDE project: on-line monitoring and
simulation validation with the in-beam PET scanner
V Ferrero1,2 on behalf of the INSIDE collaboration1 Department of Physics, University of Torino, Via Pietro Giuria 1, Torino, Italy2 Istituto Nazionale di Fisica Nucleare INFN, Sezione di Torino, Via Pietro Giuria 1, Torino,Italy
E-mail: [email protected]
Abstract. The quality assurance of particle therapy treatment is a fundamental issue thatcan be addressed by developing reliable monitoring techniques and indicators of the treatmentplan accuracy. Monitoring using Position Emission Tomography (PET) systems is the onlyin-vivo non invasive technique employed clinically and has been carried out in particle therapysince 1997. However, the PET monitoring of β+ emitter isotopes is typically done after thetreatment, resulting in a large fraction of lost data because of the isotopes rapid physical decay.The INSIDE collaboration has recently installed an in-beam PET scanner at the Italian NationalCenter of Oncologic Hadrontherapy in Pavia, Italy. Here, there is an ongoing project in orderto start testing the method on patients. This work focuses on the online performances of thescanner with clinical beams.
1. IntroductionParticle range verification is one of the key issues for treatment quality assessment inhadrontherapy, and requires the development of reliable monitoring techniques and indicators ofthe treatment plan accuracy. Charged particles release all of their energy inside the body, witha sharp ionization peak at the end of their range called the Bragg Peak (BP), so the only non-invasive way to verify the treatment accuracy is to collect the secondary radiation generatedby the beam-tissue interactions. Positron Emission Tomography (PET) has been employedclinically since 1997 [1], and offers the possibility to directly monitor the spatial distributionof positron (β+) emitter isotopes created during irradiation. Specifically, the emitted positronannihilates with an electron in the tissue, producing two back-to-back 511 keV photons, which inturn can interact and release part (i.e., Compton scattering) or all (photoelectric effect) of theirenergy in the detection crystals. Two detectors triggering in time coincidence allow the definitionof the line of response associated to a certain annihilation event. To correctly identify the linesof response the scattered events need to be discarded, so only the 511 keV events (correlated tothe photoelectric effect) are selected. The reconstructed isotopes activity distribution inside thepatient thus gives an insight about the delivered treatment.
Defining the particle range from the activity measurement is not straightforward, becauseit has a non-trivial dependency on the ion species, the energy, and the tissue composition.However, from the activity distribution it is possible to infer the BP position since its profilealong the beam path has a characteristic fall-off a few millimeters before the BP itself [2, 3, 4].
1
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distributionof this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
The INSIDE project: on-line monitoring and
simulation validation with the in-beam PET scanner
V Ferrero1,2 on behalf of the INSIDE collaboration1 Department of Physics, University of Torino, Via Pietro Giuria 1, Torino, Italy2 Istituto Nazionale di Fisica Nucleare INFN, Sezione di Torino, Via Pietro Giuria 1, Torino,Italy
E-mail: [email protected]
Abstract. The quality assurance of particle therapy treatment is a fundamental issue thatcan be addressed by developing reliable monitoring techniques and indicators of the treatmentplan accuracy. Monitoring using Position Emission Tomography (PET) systems is the onlyin-vivo non invasive technique employed clinically and has been carried out in particle therapysince 1997. However, the PET monitoring of β+ emitter isotopes is typically done after thetreatment, resulting in a large fraction of lost data because of the isotopes rapid physical decay.The INSIDE collaboration has recently installed an in-beam PET scanner at the Italian NationalCenter of Oncologic Hadrontherapy in Pavia, Italy. Here, there is an ongoing project in orderto start testing the method on patients. This work focuses on the online performances of thescanner with clinical beams.
1. IntroductionParticle range verification is one of the key issues for treatment quality assessment inhadrontherapy, and requires the development of reliable monitoring techniques and indicators ofthe treatment plan accuracy. Charged particles release all of their energy inside the body, witha sharp ionization peak at the end of their range called the Bragg Peak (BP), so the only non-invasive way to verify the treatment accuracy is to collect the secondary radiation generatedby the beam-tissue interactions. Positron Emission Tomography (PET) has been employedclinically since 1997 [1], and offers the possibility to directly monitor the spatial distributionof positron (β+) emitter isotopes created during irradiation. Specifically, the emitted positronannihilates with an electron in the tissue, producing two back-to-back 511 keV photons, which inturn can interact and release part (i.e., Compton scattering) or all (photoelectric effect) of theirenergy in the detection crystals. Two detectors triggering in time coincidence allow the definitionof the line of response associated to a certain annihilation event. To correctly identify the linesof response the scattered events need to be discarded, so only the 511 keV events (correlated tothe photoelectric effect) are selected. The reconstructed isotopes activity distribution inside thepatient thus gives an insight about the delivered treatment.
Defining the particle range from the activity measurement is not straightforward, becauseit has a non-trivial dependency on the ion species, the energy, and the tissue composition.However, from the activity distribution it is possible to infer the BP position since its profilealong the beam path has a characteristic fall-off a few millimeters before the BP itself [2, 3, 4].
JPCS
IPRD, Siena (talk)
YRM, Torino (talk)
COINCIDENCE TIME DISTRIBUTIOND
in-spill CTDinter-spill CTD
CTD [ns]
THE INSIDE PROJECT:IN-BEAM PET SCANNER SYSTEM FEATURES AND CHARACTERIZATION (14TH IPRD, SIENA)
[email protected], INFN and University of Torino
in-spill inter-spill
BEAM
REAL TREATMENT PLAN ON 15x15x20 cm3 PMMA PHANTOM
PROTON BEAM E = 83-150 MeV/u
390 ps σ
THE INSIDE PROJECT:IN-BEAM PET SCANNER SYSTEM FEATURES AND CHARACTERIZATION (14TH IPRD, SIENA)
[email protected], INFN and University of Torino
Real time data acquisition and reconstruction at
different time intervals
data 1 data 2
40 s
80 s
120 s
160 s
200 s
t (s)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
COINCIDENCES ANALYSIS
[email protected], INFN and University of Torino
data 1 data 2
40 s
80 s
120 s
160 s
200 s
t (s)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
➤ Can be launched as soon as the server saves LOR files ➤ Finds and analyses coincidences over a certain time window ➤ Can apply time selection ➤ Produces a listmode coincidences file compatible with the
reconstruction programC++, ROOT/BOOST LIBRARIES
THE INSIDE PROJECT:ONLINE MONITORING AND SIMULATION VALIDATION WITH THE IN-BEAM PET SCANNER (7TH YRM, TORINO)
[email protected], INFN and University of Torino
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
3
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
Figure 1. Left: the INSIDE in-beam PET installed in one of the CNAO treatment rooms.Right: schematic picture of the experimental set-up. The beam direction is parallel to thepositive z axis.
The total number of particles delivered per treatment was 38 × 109, resulting in about 1 Gyof dose. The energy ranges for each plan are reported in Tab. 1. The maximum single eventrate in the photopeak window was 13 MHz, with a coincidence time window set at 2 ns. Theirradiation lasted about 4 min, but data were also acquired in the after-treatment for aboutanother 6 min.
Table 1. Range energies for two real Treatment Plans (TP), protons.
TP Emin (MeV/u) Emax (MeV/u)
A 83 150B 62 129
4. SimulationsSince the distributions of β+ radioactivity and dose are not directly comparable, the qualityassessment of charged particle therapy requires the prediction of the β+ activity distributionfrom the treatment planning.
A dedicated simulation has been developed in FLUKA [11] and it comprises two steps. First,the primary beam is simulated with reduced statistics (about 1/100 of the primaries released inthe whole treatment, thus reducing the simulation time of about 70x), and the 3D β+ emitterdistribution is scored. Then, a full statistic Monte Carlo simulation is performed by extractingthe emission time of all positrons.
The simulation accounts for both the INSIDE in-beam PET characteristics and Figs. of meritand the delivery informations coming from the CNAO Dose Delivery System (DDS) [12]. TheDDS is integrated in the CNAO synchrotron and monitors the space and time structure of thebeam for each delivery. The average beam structure in time is 1 s beam on (in-spill) followedby 3-4 s beam off (inter-spill). The CNAO beam pipe and nozzle have a known geometry andhave been simulated in previous works [13].
A C++ tool is developed for post-processing the FLUKA events and obtaining a time-basedsignal. The simulated and experimental data are analyzed with the same parameters (e.g. equalcoincidence time window, energy window, and acquisition time).
MONTE CARLO SIMULATIONS
the temporal and spatial structure of the beam is simulated
isotopes decays are simulated
time
the detector characteristics are simulated
STEP 1 beam simulation
STEP 2 PET simulation
time-tagged activity scoring
image reconstruction
about 1/100 of primary hadrons
are simulated
isotope production map
all positrons are simulated (x100)
same analysis chain as real data
BEAM
BEAM
THE INSIDE PROJECT:ONLINE MONITORING AND SIMULATION VALIDATION WITH THE IN-BEAM PET SCANNER (7TH YRM, TORINO)
[email protected], INFN and University of Torino
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
3
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
Figure 1. Left: the INSIDE in-beam PET installed in one of the CNAO treatment rooms.Right: schematic picture of the experimental set-up. The beam direction is parallel to thepositive z axis.
The total number of particles delivered per treatment was 38 × 109, resulting in about 1 Gyof dose. The energy ranges for each plan are reported in Tab. 1. The maximum single eventrate in the photopeak window was 13 MHz, with a coincidence time window set at 2 ns. Theirradiation lasted about 4 min, but data were also acquired in the after-treatment for aboutanother 6 min.
Table 1. Range energies for two real Treatment Plans (TP), protons.
TP Emin (MeV/u) Emax (MeV/u)
A 83 150B 62 129
4. SimulationsSince the distributions of β+ radioactivity and dose are not directly comparable, the qualityassessment of charged particle therapy requires the prediction of the β+ activity distributionfrom the treatment planning.
A dedicated simulation has been developed in FLUKA [11] and it comprises two steps. First,the primary beam is simulated with reduced statistics (about 1/100 of the primaries released inthe whole treatment, thus reducing the simulation time of about 70x), and the 3D β+ emitterdistribution is scored. Then, a full statistic Monte Carlo simulation is performed by extractingthe emission time of all positrons.
The simulation accounts for both the INSIDE in-beam PET characteristics and Figs. of meritand the delivery informations coming from the CNAO Dose Delivery System (DDS) [12]. TheDDS is integrated in the CNAO synchrotron and monitors the space and time structure of thebeam for each delivery. The average beam structure in time is 1 s beam on (in-spill) followedby 3-4 s beam off (inter-spill). The CNAO beam pipe and nozzle have a known geometry andhave been simulated in previous works [13].
A C++ tool is developed for post-processing the FLUKA events and obtaining a time-basedsignal. The simulated and experimental data are analyzed with the same parameters (e.g. equalcoincidence time window, energy window, and acquisition time).
MONTE CARLO SIMULATIONS
time
STEP 1 beam simulation
STEP 2 PET simulation
time-tagged activity scoring
image reconstruction
about 1/100 of primary hadrons
are simulated
isotope production map
all positrons are simulated (x100)
same analysis chain as real data
HOW DO WE COMPARE THEM?the temporal and spatial structure of the beam is simulated
isotopes decays are simulated
the detector characteristics are simulated
BEAM
BEAM
ACTIVITY PROFILE
[email protected], INFN and University of Torino
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
➤ Can compare several images ➤ Image profile on preferred direction in a selectable sub-
image ➤ Calculates range with sigmoid or gaussian function in the
interval specified by the user C++, ROOT/BOOST/ITK LIBRARIES
BEAM
BEAM
BEAM
BEAM
THE INSIDE PROJECT:ONLINE MONITORING AND SIMULATION VALIDATION WITH THE IN-BEAM PET SCANNER (7TH YRM, TORINO)
[email protected], INFN and University of Torino
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
3
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
Figure 1. Left: the INSIDE in-beam PET installed in one of the CNAO treatment rooms.Right: schematic picture of the experimental set-up. The beam direction is parallel to thepositive z axis.
The total number of particles delivered per treatment was 38 × 109, resulting in about 1 Gyof dose. The energy ranges for each plan are reported in Tab. 1. The maximum single eventrate in the photopeak window was 13 MHz, with a coincidence time window set at 2 ns. Theirradiation lasted about 4 min, but data were also acquired in the after-treatment for aboutanother 6 min.
Table 1. Range energies for two real Treatment Plans (TP), protons.
TP Emin (MeV/u) Emax (MeV/u)
A 83 150B 62 129
4. SimulationsSince the distributions of β+ radioactivity and dose are not directly comparable, the qualityassessment of charged particle therapy requires the prediction of the β+ activity distributionfrom the treatment planning.
A dedicated simulation has been developed in FLUKA [11] and it comprises two steps. First,the primary beam is simulated with reduced statistics (about 1/100 of the primaries released inthe whole treatment, thus reducing the simulation time of about 70x), and the 3D β+ emitterdistribution is scored. Then, a full statistic Monte Carlo simulation is performed by extractingthe emission time of all positrons.
The simulation accounts for both the INSIDE in-beam PET characteristics and Figs. of meritand the delivery informations coming from the CNAO Dose Delivery System (DDS) [12]. TheDDS is integrated in the CNAO synchrotron and monitors the space and time structure of thebeam for each delivery. The average beam structure in time is 1 s beam on (in-spill) followedby 3-4 s beam off (inter-spill). The CNAO beam pipe and nozzle have a known geometry andhave been simulated in previous works [13].
A C++ tool is developed for post-processing the FLUKA events and obtaining a time-basedsignal. The simulated and experimental data are analyzed with the same parameters (e.g. equalcoincidence time window, energy window, and acquisition time).
BEAM
BEAM
ACTIVITY COMPARISON
[email protected], INFN and University of Torino
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
➤ Compares two images ➤ Can be launched in time (bash script) ➤ Selects the activity through the application of image
filtering and compares the images 3-dimentionally ➤ Can use the treatment plan spatial information to select a
sub-image of activity
C++, ROOT/BOOST/ITK LIBRARIESBE
AM
BEAM
THE INSIDE PROJECT:ONLINE MONITORING AND SIMULATION VALIDATION WITH THE IN-BEAM PET SCANNER (7TH YRM, TORINO)
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
ACTIVITY SELECTION
TP A (data) TP A (sim) TP B (data) TP B (sim)
BEAM’S EYE VIEW
3
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
Figure 1. Left: the INSIDE in-beam PET installed in one of the CNAO treatment rooms.Right: schematic picture of the experimental set-up. The beam direction is parallel to thepositive z axis.
The total number of particles delivered per treatment was 38 × 109, resulting in about 1 Gyof dose. The energy ranges for each plan are reported in Tab. 1. The maximum single eventrate in the photopeak window was 13 MHz, with a coincidence time window set at 2 ns. Theirradiation lasted about 4 min, but data were also acquired in the after-treatment for aboutanother 6 min.
Table 1. Range energies for two real Treatment Plans (TP), protons.
TP Emin (MeV/u) Emax (MeV/u)
A 83 150B 62 129
4. SimulationsSince the distributions of β+ radioactivity and dose are not directly comparable, the qualityassessment of charged particle therapy requires the prediction of the β+ activity distributionfrom the treatment planning.
A dedicated simulation has been developed in FLUKA [11] and it comprises two steps. First,the primary beam is simulated with reduced statistics (about 1/100 of the primaries released inthe whole treatment, thus reducing the simulation time of about 70x), and the 3D β+ emitterdistribution is scored. Then, a full statistic Monte Carlo simulation is performed by extractingthe emission time of all positrons.
The simulation accounts for both the INSIDE in-beam PET characteristics and Figs. of meritand the delivery informations coming from the CNAO Dose Delivery System (DDS) [12]. TheDDS is integrated in the CNAO synchrotron and monitors the space and time structure of thebeam for each delivery. The average beam structure in time is 1 s beam on (in-spill) followedby 3-4 s beam off (inter-spill). The CNAO beam pipe and nozzle have a known geometry andhave been simulated in previous works [13].
A C++ tool is developed for post-processing the FLUKA events and obtaining a time-basedsignal. The simulated and experimental data are analyzed with the same parameters (e.g. equalcoincidence time window, energy window, and acquisition time).
OVERALL VIEW
FWHM = 3.2 mm FWHM = 4.8 mm
FWHM = 3.2 mm FWHM = 3.2 mm
TP A 200s (data vs sim)
TP B 200s (data vs sim)
BEAM
BEAM
THE INSIDE PROJECT:ONLINE MONITORING AND SIMULATION VALIDATION WITH THE IN-BEAM PET SCANNER (7TH YRM, TORINO)
TP A (data)
40 s
80 s
120 s
160 s
200 s
t (s)
TP A (sim)
0
20
10
0
80
40
20
60
0
120
60
30
90
0
200
100
50
150
5
15
0
300
200
100
TP B (data) TP B (sim)
BEAM’S EYE VIEW
3
1234567890
7th Young Researcher Meeting IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 841 (2017) 012011 doi :10.1088/1742-6596/841/1/012011
Figure 1. Left: the INSIDE in-beam PET installed in one of the CNAO treatment rooms.Right: schematic picture of the experimental set-up. The beam direction is parallel to thepositive z axis.
The total number of particles delivered per treatment was 38 × 109, resulting in about 1 Gyof dose. The energy ranges for each plan are reported in Tab. 1. The maximum single eventrate in the photopeak window was 13 MHz, with a coincidence time window set at 2 ns. Theirradiation lasted about 4 min, but data were also acquired in the after-treatment for aboutanother 6 min.
Table 1. Range energies for two real Treatment Plans (TP), protons.
TP Emin (MeV/u) Emax (MeV/u)
A 83 150B 62 129
4. SimulationsSince the distributions of β+ radioactivity and dose are not directly comparable, the qualityassessment of charged particle therapy requires the prediction of the β+ activity distributionfrom the treatment planning.
A dedicated simulation has been developed in FLUKA [11] and it comprises two steps. First,the primary beam is simulated with reduced statistics (about 1/100 of the primaries released inthe whole treatment, thus reducing the simulation time of about 70x), and the 3D β+ emitterdistribution is scored. Then, a full statistic Monte Carlo simulation is performed by extractingthe emission time of all positrons.
The simulation accounts for both the INSIDE in-beam PET characteristics and Figs. of meritand the delivery informations coming from the CNAO Dose Delivery System (DDS) [12]. TheDDS is integrated in the CNAO synchrotron and monitors the space and time structure of thebeam for each delivery. The average beam structure in time is 1 s beam on (in-spill) followedby 3-4 s beam off (inter-spill). The CNAO beam pipe and nozzle have a known geometry andhave been simulated in previous works [13].
A C++ tool is developed for post-processing the FLUKA events and obtaining a time-basedsignal. The simulated and experimental data are analyzed with the same parameters (e.g. equalcoincidence time window, energy window, and acquisition time).
OVERALL VIEW
TP A (data vs sim)
TP A (data vs sim) TP B (data vs sim)
TP B (data vs sim)
FWHM (1.8 - 6.2) mm
FWHM (1.5 - 3.2) mm
FWHM (2.2 - 6.0) mm
FWHM (3.2 - 4.8) mm
[email protected], INFN and University of Torino
[-1.2, 1.9] mm after 120 s
[-0.1, 0.8] mm after 120 s
END OF TREATMENT
* *
* *BE
AM
BEAM
DOUBLE FIELD HADRONTHERAPY TREATMENT MONITORING WITH THE INSIDE IN-BEAM PET: PROOF OF CONCEPT ON PMMA PHANTOMS (IEEE NSS/MIC, ATLANTA, USA)
21
BEAM
[email protected], INFN and University of Torino
A novel procedure to monitor multi-field hadrontherapy treatments with in-beam PET: proof of concept on PMMA phantoms
Ferrero V.1,2, on behalf of the INSIDE collaboration [1] University of Torino, [2] INFN Torino
Positron Emission Tomography (PET) is a reliable imaging technique for in-vivo monitoring in hadrontherapy treatment sessions. Exploiting the fragmentation processes in the patient tissues and in the beam projectile (if carbon ion), the produced β+ activity can be measured by collecting the back-to-back 511 keV photons coming from positron annihilation. In-beam PET, in particular, allows monitoring while the treatment is ongoing, minimizing the biological washout effects and displaying the highest correlation between the measured activity and the deposited dose [1]. During 2016 the INSIDE (Innovative Solutions for In-beam Dosimetry in hadronthErapy) in-beam PET was installed at the Italian Center of Oncological Hadrontherpy (CNAO) in Pavia, Italy, and characterized with PMMA phantoms [2-5]. In December 2016, the first patient was monitored in-vivo. Given the promising results, it was decided to upgrade the mechanical support of the detector so as to allow the monitoring of the majority of patients and their different beam fields. Multiple fields are standard protocol in radiation therapy to deliver a more conformal dose and better spare normal tissues and organ at risk. There are currently no relevant results on monitoring the cumulative activity of more than one beam field [6-7], because of obvious reasons (i.e., washout processes, residual activity from previous irradiations). Still, discerning the activity of subsequent beam fields from previous ones would allow quality control not for just one beam field, but for the whole treatment session. Two treatment plans, one with carbon ions, the other with protons, consisting of two beam fields each, were delivered on PMMA phantoms and monitored with the INSIDE in-beam PET scanner. The results are here presented.
The INSIDE in-beam PET scanner. The scanner is based on two planar heads consisting on 2x5 detection modules of 16x16 segmented lutetium fine silicate scintillating crystals, coupled 1:1 to Hamamatsu MPPCs, resulting in 25602 lines of response within the Field Of View (FOV). The 64 channel TOFPET ASIC [8] is used to process information about the energy and detection time of each event. A FPGA-based data processing system selects the 511 keV events and transmits them to a custom Data Acquisition System (DAQ) implemented on a server (32 Hyper Thread cores and 128 GB RAM) in which a high performance software applies on-line data sorting and writes time-tagged coincidence data. A maximum likelihood expectation maximization algorithm with 5 iterations is used to reconstruct time-resolved activity images (FOV 224×112×264 mm3, with voxel size 1.6x1.6x1.6 mm3). The detailed description of the system is reported in [2].
Experimental set-up. The in-beam PET scanner was positioned through manual alignment with the treatment room lasers. The working distance between the heads was set at 50 cm, with the FOV at the isocenter. Two treatment plans were selected from the CNAO database and delivered to two 15x15x20 cm3 PMMA homogeneous phantoms. Both plans consisted of two beam fields, the second of which was delivered on the same phantom, after a rotation of 180° around the isocenter, so that the first beam field residual activity resulted superimposed to the second in the distal fall-off area.
The delivered treatments details are reported in table 1.
Table 1. Energy ranges of the delivered treatments.
Data acquisition. The CNAO synchrotron presents a pulsed beam structure comprising actual beam delivery (in-spill) followed by beam off (inter-spill). At the moment of the acquisition, the in/inter-spill ratio was 1:2 s. The DAQ system allows on-line selection of in/inter-spill coincidences, consenting real time reconstruction of both data sets. Because of prompt radiation and neutron background during the spill though, only inter-spill data are considered significant for the analysis. Figure 1 shows the coincidence event rate for the proton treatment delivery. The in-spill phase is shown in red; the inter-spill in blue. After the first field irradiation, the phantom is repositioned and the second beam field is delivered. The maximum coincidence event rate detected for the proton and carbon ion treatment plan was of 18 and 3 kHz, respectively, with a coincidence time window set at 2 ns.
# Figure 1. Coincidence event rate of a two-field treatment plan, protons. The two fields time-intervals are highlighted in the figure, along with the repositioning of the phantom. The red spikes correspond to the spill phase, while the inter-spill is shown in blue. Results. Specific intervals of time-tagged acquired data can be selected through a custom data analysis software and reconstructed at different irradiation moments. This allows the identification of the first field alone, followed by the second field superimposed to the first field long decaying isotopes. In figure 2 and 3 the proton and carbon ion treatment plans reconstructed at different time intervals are reported. The time intervals are selected so that the first field is shown before and after the repositioning of the phantom. The following irradiation selection starts from the beginning of the second field, so as to have as less residual activity as possible, and comprises 145 s aftertreatment in both cases. A median filter with a 5x5x5 mm3 kernel was applied to the PET images because of low statistic (especially for the carbon ion case) and shot noise.
Treatment Emin (MeV/u) Emax (MeV/u)
Proton beamField 1 83 150
Field 2 62 129
Carbon ion beamField 1 134 269
Field 2 134 264
Field 1 Field 2 Total Dose
180°
Coincidences event rate, proton beam
f(t) = C e�t/⌧
⌧ = 274 s
!Z t1
t0
f(t)dt
t [s]
DOUBLE FIELD HADRONTHERAPY TREATMENT MONITORING WITH THE INSIDE IN-BEAM PET: PROOF OF CONCEPT ON PMMA PHANTOMS (IEEE NSS/MIC, ATLANTA, USA)
[email protected], INFN and University of Torino
IMAGE COMPARISON
BEAM
’S EY
E VIE
WOV
ERAL
L VIE
W
FWHM = 6.4 mm
FWHM = 1.6 mm
PROT
ON B
EAM
TREA
TMEN
T PLA
N
Coincidence event rate
DOUBLE FIELD HADRONTHERAPY TREATMENT MONITORING WITH THE INSIDE IN-BEAM PET: PROOF OF CONCEPT ON PMMA PHANTOMS (IEEE NSS/MIC, ATLANTA, USA)
[email protected], INFN and University of Torino
IMAGE COMPARISON
BEAM
’S EY
E VIE
WOV
ERAL
L VIE
W
FWHM = 11.2 mm
FWHM = 4.8 mm
f(t) = Ce�t/⌧
⌧ = 250s
!Z t1
t0
f(t)dt
CARB
ON IO
N BE
AM TR
EATM
ENT P
LAN
Coincidence event rate
FIRST CLINICAL TEST
[email protected], INFN and University of Torino
INFN - COMMUNICATIONS OFFICE 2
PHYSICS AND MEDICINECANCER: INSIDE SYSTEM SUCCESSFULLY TESTED FOR THE FIRST TIME ON A PATIENT
INSIDE (Innovative Solutions for Dosimetry in Hadrontherapy) has been tested for the first time on a patient. This innovative imaging system, which uses particle accelerators, was built by the INFN in
Turin to further enhance the efficacy of hadron therapy, used for the treatment of localised tumours. INSIDE, which received a € 1 million grant under the PRIN (Relevant National Interest Projects) program, is the result of a research project coordinated by the University of Pisa in collaboration with the Universities the universities of Turin and Sapienza of Rome, Bari Polytechnic University and INFN. For the trial phase, INSIDE was tested on the patient at the Italian National Centre for Oncological Hadron Therapy (CNAO), in Pavia. INSIDE is an innovative monitoring system, which uses detector technology to obtain images of what happens inside the patient's body during the hadron therapy treatment. In more detail, this bimodal imaging system combines a positron emission tomography (PET) scanner with a tracking system for charged particle imaging and is capable of operating during radiation delivery to treat head and neck tumours. ▪
NEWSLETTER 30Italian National Institute for Nuclear Physics
DECEMBER 2016
FIRST CLINICAL TEST
[email protected], INFN and University of Torino
End of treatment
Data 1/12/16
Data 2/12/16
Simulation 1/12/16
Comparison analysis of the experimental data acquired in the two consecutive days: Ferrero V., Fiorina E., Morrocchi M., Pennazio F. et al, Online proton therapy monitoring: clinical test of a Silicon-photodetector-based in-beam PET, submitted to Nature Scientific Reports
BEAM’S EYE VIEW
OVERALL VIEW
IN THE NEXT FUTURE: RECONSTRUCTION ALGORITHM
27
Simulation of a 10x10x12 cm3 water phantom at different positions along z (beam axis) 30 kBq activity (18F-FDG) Maximum Likelihood Expectation Maximization (MLEM) reconstruction model with 5 iteration
right side in FOV central point in FOV left side in FOV
slice in x40 50 60 70 80 90 100
z[cm
]
25
25.2
25.4
25.6
25.8
26
26.2
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Exit inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
12.8
13
13.2
13.4
13.6
13.8
14
14.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Entrance inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
25
25.2
25.4
25.6
25.8
26
26.2
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Exit inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
12.8
13
13.2
13.4
13.6
13.8
14
14.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Entrance inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
7
7.2
7.4
7.6
7.8
8
8.2
8.4slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Entrance inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
18.8
19
19.2
19.4
19.6
19.8
20
20.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Exit inflection point
[email protected], INFN and University of Torino
The reconstruction algorithm is limited by the PET dual head geometry, but it could be optimized, e.g, by integrating the beam information → Visiting at Lubëck University, Germany, 2018
slice in x40 50 60 70 80 90 100
z[cm
]
12.8
13
13.2
13.4
13.6
13.8
14
14.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Exit inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
0.8
1
1.2
1.4
1.6
1.8
2
2.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Entrance inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
12.8
13
13.2
13.4
13.6
13.8
14
14.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Exit inflection point
slice in x40 50 60 70 80 90 100
z[cm
]
0.8
1
1.2
1.4
1.6
1.8
2
2.2slice in y
6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
slice in y6121824303642485460
Entrance inflection point
IN THE NEXT FUTURE➤ Mechanical cart upgrade
➤ PET/tracker combined acquisition tests
➤ Longitudinal clinical trials @ CNAO
➤ Image reconstruction algorithm: visiting period (3 months - Erasmus Traineeship grant) @ Lubëck University, Germany
[email protected], INFN and University of Torino
CONFERENCE AND WORKSHOP PRESENTATIONS
29
➤ IEEE NSS/MIC, Atlanta, Georgia, USA
➤ Fisica e Informatica in Medicina, Monza, Italy
➤ 7th Young Researchers Meeting, Torino, Italy
➤ 14th Topical Seminar on Innovative Particle and Radiation Detectors, Siena, Italy
➤ 102° Italian National Congress, Padua, Italy
2016
2017
PAPERS➤ The INSIDE project: in-beam pet scanner system features and characterization. Journal of Instrumentation, 12(03):C03051,
2017.
➤ The INSIDE project: on-line monitoring and simulation validation with the in-beam PET scanner. Journal of Physics: Conference Series, 841(1):012011, 2017.
➤ INSIDE in-beam positron emission tomography system for particle range monitoring in hadrontherapy. Journal of Medical Imaging, 4(1), 2016.
➤ Full-beam performances of a pet detector with synchrotron therapeutic proton beams. Physics in Medicine and Biology, 61(23):N650–N666, 2016.
➤ First results of the INSIDE in-beam pet scanner for the on-line monitoring of particle therapy treatments. Journal of Instrumentation, 11(12):C12011, 2016.
PUBL
ISH
ED
SUBMITTED Online proton therapy monitoring: clinical test of a Silicon-photodetector-based in-beam PET, Nature Scientific Reports
➤ Monte Carlo Simulations and data analysis of carbon ions beam therapy monitoring: a case study with the INSIDE in-beam PET
➤ Monte Carlo simulation tool for online treatment monitoring in hadrontherapy with in-beam PET
➤ Double-field hadrontherapy treatment monitoring with the INSIDE in-beam PET: proof of concept on PMMA phantoms
WO
RK
IN
PRO
GR
ESS