Particle Therapy (PT) is an advanced form of radiation therapy that utilizes protons, heavy ions (e.g., carbon ions), or neutrons to target tumors with high precision, minimizing toxicity to surrounding tissues compared to conventional radiotherapy. Unlike X-rays and electrons, which deposit dose exponentially with depth, heavy charged particles exhibit a Bragg peak, a sharp dose deposition at the end of their range, allowing precise targeting of tumor depths. However, PT’s sensitivity to uncertainties, such as anatomical changes, physiological organ movement, and particle range calculations, can lead to dose mismatches, often necessitating larger safety margins and multiple fields, which increases radiation exposure to surrounding tissues. In vivo particle range verification, particularly using prompt gamma imaging (PGI), has been identified as a promising solution for real-time dose profile monitoring, allowing for potential adjustments during treatment. PGI is based on prompt gamma photons emitted almost instantly by nuclear interactions within the patient's tissues. This technique was already tested in clinical environment with a knife-edge-collimator camera for proton treatments achieving a range verification precision down to 1 mm, but remains relatively unexplored for Carbon Ion Radiation Therapy (CIRT). C-ions offer several distinct advantages over protons in hadrontherapy due to their unique physical interaction properties, which result in a higher linear energy transfer (LET) and thus an increased relative biological effectiveness (RBE). This enhanced RBE allows C-ions to achieve a more effective therapeutic outcome, especially in treating hypoxic tumors, which are typically more resistant to conventional radiation therapies. However, the use of prompt gamma imaging (PGI) technique in C-ion therapy introduces specific challenges. Notably, C-ions produce a higher neutron yield per ion and require fewer incident particles to deliver the same physical dose as protons. These characteristics significantly impact the signal-to-background ratio, presenting a critical challenge for accurate PGI measurements. Consequently, the optimization of detection efficiency and the ability to discriminate between gamma and neutron emissions are paramount in adapting PGI for Carbon Ion Radiation Therapy (CIRT). In this Ph.D. thesis, the potential of PGI for range verification in CIRT environment has been explored using a prototype knife-edge slit-camera system. First, Monte Carlo simulations were conducted using FLUKA, a versatile Monte Carlo code for particle transport. In those simulations, a knife-edge slit camera was employed to detect secondary particles emitted in the 3-7 MeV energy range from an ICRP (International Commission on Radiological Protection) soft tissue phantom, simulating a patient irradiated with a mono-energetic pencil beam of carbon ions (C-ions) at 150 MeV/u. This energy was chosen to align with the therapeutic energy range used in CIRT. The simulation results indicated that a layer-by-layer range verification could be achieved with a 4mm precision using a 10 cm x 10 cm detection module, provided that an average of approximately 5 x 10^7 primary particles are delivered. To validate these findings, a first 64-ch prototype was developed and tested at the Centro Nazionale di Adronterapia Oncologica (CNAO) in Pavia, Italy. This experiment employed a knife-edge-collimator camera prototype, utilizing a 8×8 pixelated LYSO scintillator (5 x 5 cm^2) coupled with a 64-SiPM array, four GAMMA (Gain Amplitude Modulated Multichannel ASIC) ASICs and a FPGA-based DAQ system. Prompt gamma (PG) profiles were measured by irradiating a plastic phantom with a C-ion pencil beam at clinical energies and intensities. Additionally, the detector was translated in different positions to expand the field of view (FOV) to 13 x 5 cm^2. The prototype successfully detected Bragg-peak shifts with a precision of approximately 4 mm for a sample statistic of roughly 4x10^8 C-ions (3x10^8 for the extended FOV), which was larger than initially anticipated. Despite this, the detector demonstrated significant potential for accurately verifying dose delivery precision after a treatment fraction, which is essential in clinical practice. To our knowledge, this experiment marks the first instance where range verification based on PGI was applied to a carbon ion beam at clinical energy and intensities, highlighting the viability of PGI for real-time range verification in C-ion therapy. A second 32-ch prototype featuring a 8×4 pixelated LYSO scintillator (5 x 2.5 cm^2) coupled with a 32-SiPM array, two SITH (Spectroscopy Imaging Timing Hadrontherapy) ASICs and a FPGA-based DAQ system. This system has better count-rate capabilities, up to 700kHz per channel and timing capabilities (less than 1ns). Recently, also this system has been tested at CNAO both with carbon ion and proton beams and the preliminary results indicated the possibility to detect the PG shifts at high statistics for both the particle beams. Moreover, for the particularly challenging extraction of the prompt gamma profile—correlated to the Bragg peak fall-off position—from the neutron background, Pulse Shape Discrimination (PSD) was investigated using crystals with selective responses to neutrons and gammas. A first prototype was developed based on a 1"x1" CLYC-6 (Cesium Lithium Yttrium Chlorid) scintillator coupled with 4x4 array of SiPMs and read out by a SITH ASIC. This prototype was tested at CNAO during the experimental campaign using the 64-ch PGI detector prototype, using the same irradiation setup. The Figure of Merit (FOM) obtained was 1.29 at t_1 = 120 ns and t_2 = 1500 ns, which is adequate to effectively perform PSD. This result provides a further option for future research on PGI, based on the CLYC scintillator, particularly when paired with the SITH ASIC, to mitigate neutron background interference in prompt gamma-ray (PG) measurements during hadrontherapy.
La Terapia con Particelle (TP) è una forma avanzata di radioterapia che utilizza protoni, ioni pesanti (come gli ioni carbonio) o neutroni per colpire i tumori con elevata precisione, riducendo al minimo la tossicità per i tessuti circostanti rispetto alla radioterapia convenzionale. A differenza dei raggi X e degli elettroni, che depositano la dose in modo esponenziale con la profondità, le particelle cariche pesanti mostrano un picco di Bragg, ossia una deposizione di dose molto precisa alla fine della loro traiettoria, consentendo un mirato trattamento in profondità del tumore. Tuttavia, la PT è sensibile a incertezze come cambiamenti anatomici, movimento fisiologico degli organi e calcoli del range delle particelle, il che può causare disallineamenti di dose, richiedendo spesso margini di sicurezza maggiori e campi multipli, che aumentano l’esposizione alle radiazioni per i tessuti circostanti. La verifica in vivo del range di particelle, in particolare tramite la tecnica di imaging dei gamma pronti (PGI, Prompt Gamma Imaging), è stata identificata come una soluzione promettente per il monitoraggio in tempo reale del profilo di dose, consentendo potenziali aggiustamenti durante il trattamento. La PGI si basa sui fotoni gamma pronti emessi quasi istantaneamente dalle interazioni nucleari nei tessuti del paziente. Questa tecnica è già stata testata in ambiente clinico tramite una camera collimata a knife-edge per trattamenti protonici, raggiungendo una precisione di verifica del range fino a 1mm, ma rimane relativamente inesplorata per la Terapia con Ioni Carbonio (CIRT, Carbon Ion Radiation Therapy). Gli ioni carbonio presentano diversi vantaggi rispetto ai protoni nella adroterapia grazie alle loro proprietà uniche di interazione fisica, che comportano un maggiore trasferimento lineare di energia (LET, Linear Energy Transfer) e quindi una più alta efficacia biologica relativa (RBE, Relative Biological Effectiveness). Questo RBE migliorato consente agli ioni carbonio di ottenere un risultato terapeutico più efficace, specialmente nel trattamento di tumori ipossici, che sono generalmente più resistenti alle terapie radiologiche convenzionali. Tuttavia, l'uso della tecnica PGI nella terapia con ioni carbonio introduce delle sfide. In particolare, gli ioni carbonio producono una maggiore quantità di neutroni per particella e richiedono meno particelle incidenti per fornire la stessa dose fisica dei protoni. Queste caratteristiche influenzano significativamente il rapporto segnale/rumore, rappresentando una sfida per misure PGI. Di conseguenza, l'ottimizzazione dell'efficienza di rilevamento e la capacità di discriminare tra emissioni di gamma e neutroni sono fondamentali per adattare la PGI alla CIRT. In questa tesi di dottorato, il potenziale della PGI per la verifica del range nella CIRT è stato esplorato utilizzando un prototipo di camera a knife-edge. Per prima cosa, sono state effettuate simulazioni Monte Carlo utilizzando FLUKA, un versatile codice Monte Carlo per il trasporto di particelle. In queste simulazioni, è stata utilizzata una camera knife-edge per rilevare particelle secondarie emesse nel range di energia 3-7MeV da un fantoccio di tessuto molle ICRP, simulando un paziente irradiato con un fascio di ioni di carbonio monoenergetico a 150MeV/u. Questa energia è stata scelta per allinearsi con l’intervallo terapeutico utilizzato nella CIRT. I risultati delle simulazioni indicano che una verifica del range a livello di uno spill potrebbe essere raggiunta con una precisione di 4mm utilizzando un rivelatore di 10cm x 10cm, a condizione che venga somministrato un numero medio di circa 5 x 10^7 particelle primarie. Per validare questi risultati, è stato sviluppato e testato un primo prototipo a 64 canali presso il Centro Nazionale di Adroterapia Oncologica (CNAO) a Pavia, in Italia. Questo esperimento ha utilizzato un prototipo di camera collimata a knife-edge, dotato di uno scintillatore LYSO pixelato 8x8 (5 x 5\,cm^2) accoppiato a una matrice di 64 SiPM, quattro ASIC GAMMA (Gain Amplitude Modulated Multichannel ASIC) e un sistema di acquisizione dati basato su FPGA. I profili gamma pronti sono stati misurati irradiando un fantoccio di plastica con un fascio di ioni carbonio a energie e intensità cliniche. Inoltre, il rivelatore è stato traslato in diverse posizioni per espandere il campo visivo a 13 x 5cm^2. Il prototipo ha rilevato con successo spostamenti del picco di Bragg con una precisione di circa 4\,mm per un campione statistico di circa 4 x 10^8 ioni carbonio (3 x 10^8 per il campo visivo esteso), maggiore di quanto inizialmente previsto. Tuttavia, il rivelatore ha dimostrato un significativo potenziale per la verifica della dose dopo una frazione di trattamento, essenziale nella pratica clinica. A nostra conoscenza, questo esperimento rappresenta la prima volta in cui la verifica del range basata su PGI è stata applicata a un fascio di ioni carbonio con energia e intensità cliniche, evidenziando la fattibilità della PGI per la verifica in tempo reale del range nella terapia con ioni carbonio. Un secondo prototipo a 32 canali, caratterizzato da uno scintillatore LYSO pixelato 8x4 (5 x 2.5cm^2) accoppiato a una matrice di 32 SiPM, due ASIC SITH (Spectroscopy Imaging Timing Hadrontherapy) e un sistema di acquisizione dati basato su FPGA, è stato recentemente testato anch’esso presso CNAO sia con fasci di ioni carbonio che di protoni. I risultati preliminari hanno indicato la possibilità di rilevare gli spostamenti delle curve PG a statistiche elevate per entrambi i fasci di particelle. Inoltre, per l’estrazione particolarmente complessa del profilo dei gamma pronti, correlato alla posizione del picco di Bragg, dal fondo neutronico, è stata esplorata la tecnica del Pulse Shape Discrimination (PSD) utilizzando cristalli con risposte selettive a neutroni e gamma. È stato sviluppato un primo prototipo basato su uno scintillatore CLYC-6 (Cloruro di Cesio Litio Ittrio) 1"x1" accoppiato a una matrice di SiPM 4x4 e letto tramite l'ASIC SITH. Questo prototipo è stato testato al CNAO durante la campagna sperimentale con il prototipo PGI a 64 canali, utilizzando lo stesso setup di irradiazione. La figure di merito (FOM, Figure of Merit) ottenuta è stata di 1,29 a t_1 = 120ns e t_2 = 1500ns, sufficiente per effettuare efficacemente la PSD. Questo risultato apre la strada per ulteriori ricerche sulla PGI, basate sullo scintillatore CLYC e particolarmente efficace in combinazione con l’ASIC SITH, per mitigare l’interferenza del fondo neutronico nelle misure dei gamma pronti durante l’adroterapia.
Study and development of a prompt gamma imaging detector for range monitoring in carbon ion radiation therapy
Aicha, Bourkadi Idrissi
2025
Abstract
Particle Therapy (PT) is an advanced form of radiation therapy that utilizes protons, heavy ions (e.g., carbon ions), or neutrons to target tumors with high precision, minimizing toxicity to surrounding tissues compared to conventional radiotherapy. Unlike X-rays and electrons, which deposit dose exponentially with depth, heavy charged particles exhibit a Bragg peak, a sharp dose deposition at the end of their range, allowing precise targeting of tumor depths. However, PT’s sensitivity to uncertainties, such as anatomical changes, physiological organ movement, and particle range calculations, can lead to dose mismatches, often necessitating larger safety margins and multiple fields, which increases radiation exposure to surrounding tissues. In vivo particle range verification, particularly using prompt gamma imaging (PGI), has been identified as a promising solution for real-time dose profile monitoring, allowing for potential adjustments during treatment. PGI is based on prompt gamma photons emitted almost instantly by nuclear interactions within the patient's tissues. This technique was already tested in clinical environment with a knife-edge-collimator camera for proton treatments achieving a range verification precision down to 1 mm, but remains relatively unexplored for Carbon Ion Radiation Therapy (CIRT). C-ions offer several distinct advantages over protons in hadrontherapy due to their unique physical interaction properties, which result in a higher linear energy transfer (LET) and thus an increased relative biological effectiveness (RBE). This enhanced RBE allows C-ions to achieve a more effective therapeutic outcome, especially in treating hypoxic tumors, which are typically more resistant to conventional radiation therapies. However, the use of prompt gamma imaging (PGI) technique in C-ion therapy introduces specific challenges. Notably, C-ions produce a higher neutron yield per ion and require fewer incident particles to deliver the same physical dose as protons. These characteristics significantly impact the signal-to-background ratio, presenting a critical challenge for accurate PGI measurements. Consequently, the optimization of detection efficiency and the ability to discriminate between gamma and neutron emissions are paramount in adapting PGI for Carbon Ion Radiation Therapy (CIRT). In this Ph.D. thesis, the potential of PGI for range verification in CIRT environment has been explored using a prototype knife-edge slit-camera system. First, Monte Carlo simulations were conducted using FLUKA, a versatile Monte Carlo code for particle transport. In those simulations, a knife-edge slit camera was employed to detect secondary particles emitted in the 3-7 MeV energy range from an ICRP (International Commission on Radiological Protection) soft tissue phantom, simulating a patient irradiated with a mono-energetic pencil beam of carbon ions (C-ions) at 150 MeV/u. This energy was chosen to align with the therapeutic energy range used in CIRT. The simulation results indicated that a layer-by-layer range verification could be achieved with a 4mm precision using a 10 cm x 10 cm detection module, provided that an average of approximately 5 x 10^7 primary particles are delivered. To validate these findings, a first 64-ch prototype was developed and tested at the Centro Nazionale di Adronterapia Oncologica (CNAO) in Pavia, Italy. This experiment employed a knife-edge-collimator camera prototype, utilizing a 8×8 pixelated LYSO scintillator (5 x 5 cm^2) coupled with a 64-SiPM array, four GAMMA (Gain Amplitude Modulated Multichannel ASIC) ASICs and a FPGA-based DAQ system. Prompt gamma (PG) profiles were measured by irradiating a plastic phantom with a C-ion pencil beam at clinical energies and intensities. Additionally, the detector was translated in different positions to expand the field of view (FOV) to 13 x 5 cm^2. The prototype successfully detected Bragg-peak shifts with a precision of approximately 4 mm for a sample statistic of roughly 4x10^8 C-ions (3x10^8 for the extended FOV), which was larger than initially anticipated. Despite this, the detector demonstrated significant potential for accurately verifying dose delivery precision after a treatment fraction, which is essential in clinical practice. To our knowledge, this experiment marks the first instance where range verification based on PGI was applied to a carbon ion beam at clinical energy and intensities, highlighting the viability of PGI for real-time range verification in C-ion therapy. A second 32-ch prototype featuring a 8×4 pixelated LYSO scintillator (5 x 2.5 cm^2) coupled with a 32-SiPM array, two SITH (Spectroscopy Imaging Timing Hadrontherapy) ASICs and a FPGA-based DAQ system. This system has better count-rate capabilities, up to 700kHz per channel and timing capabilities (less than 1ns). Recently, also this system has been tested at CNAO both with carbon ion and proton beams and the preliminary results indicated the possibility to detect the PG shifts at high statistics for both the particle beams. Moreover, for the particularly challenging extraction of the prompt gamma profile—correlated to the Bragg peak fall-off position—from the neutron background, Pulse Shape Discrimination (PSD) was investigated using crystals with selective responses to neutrons and gammas. A first prototype was developed based on a 1"x1" CLYC-6 (Cesium Lithium Yttrium Chlorid) scintillator coupled with 4x4 array of SiPMs and read out by a SITH ASIC. This prototype was tested at CNAO during the experimental campaign using the 64-ch PGI detector prototype, using the same irradiation setup. The Figure of Merit (FOM) obtained was 1.29 at t_1 = 120 ns and t_2 = 1500 ns, which is adequate to effectively perform PSD. This result provides a further option for future research on PGI, based on the CLYC scintillator, particularly when paired with the SITH ASIC, to mitigate neutron background interference in prompt gamma-ray (PG) measurements during hadrontherapy.File | Dimensione | Formato | |
---|---|---|---|
Politecnico_di_Milano___DEIB_PhD_Thesis_Aicha_Bourkadi_Idrissi.pdf
accesso solo da BNCF e BNCR
Dimensione
42.42 MB
Formato
Adobe PDF
|
42.42 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/207552
URN:NBN:IT:POLIMI-207552