Boron Neutron Capture Therapy (BNCT) is a radiotherapy technique that exploits the 10B(n,α)7Li reaction to kill tumor cells while sparing surrounding healthy tissues. An important aspect of this type of treatment is monitoring the dose delivery, that can be done in real time employing PG-SPECT. This work focuses on the development of BeNEdiCTE prototype detector, that aims at measuring and reconstructing the spatial distribution of the absorbed dose during BNCT treatments by the detection of the 478 keV photon, emitted in 94% of the neutron capture reactions on 10B. Monte Carlo simulations were performed on FLUKA to predict the performance of the detector and optimize its geometry. The spatial resolution of the collimator was evaluated, and subsequently improved by adding a lead patch. The optimized configuration was then implemented on FLUKA in the geometry representing the PGNAAline at LENA facility in Pavia, in order to simulate three different SPECT projection images of two borated vials, that were then used to obtain a 3D reconstruction through MLEM algorithm. The simulated projections were then compared to the results acquired during a measurement session carried out at LENA. The good agreement between simulations and experimental results demonstrates the ability of the simulations to predict the response of the detector, and its capability in performing SPECT acquisitions for tomographic reconstructions, with clinically compatible acquisition times. The performance of the detector was then simulated at the NUANS facility in Nagoya, under epithermal neutron irradiation field and high background condition. The results show that distinguishable SPECT projections can be obtained within acceptable acquisition times, providing a basis for further improvements and experimental validation.
La Boron Neutron Capture Therapy (BNCT) è una tecnica di radioterapia che sfrutta la reazione 10B(n,α)7Li per inattivare le cellule tumorali, risparmiando allo stesso tempo i tessuti sani circostanti. Un aspetto importante di questo tipo di trattamento è il monitoraggio della dose somministrata, che può essere eseguito in tempo reale tramite PG-SPECT. Questa tesi si concentra sullo sviluppo di BeNEdiCTE, un prototipo di rivelatore progettato per misurare e ricostruire la distribuzione spaziale della dose assorbita durante trattamenti BNCT, attraverso la rivelazione dei fotoni a 478 keV emessi nel 94% delle reazioni di cattura neutronica sul 10B. Sono state eseguite simulazioni Monte Carlo tramite il codice FLUKA per prevedere il comportamento del detector e ottimizzarne la geometria. La risoluzione spaziale del collimatore è stata valutata e migliorata mediante l’aggiunta di un inserto di piombo. Successivamente, la configurazione ottimizzata è stata implementata in FLUKA nella geometria rappresentate la linea PGNAA del laboratorio LENA di Pavia, al fine di simulare tre diverse proiezioni SPECT di due provette contenenti boro, utilizzate poi per una ricostruzione 3D tramite un algoritmo MLEM. Le proiezioni simulate sono state poi confrontate con i risultati ottenuti durante una sessione di misura svolta al LENA. Il buon accordo tra simulazioni e risultati sperimentali ha confermato l’abilità delle simulazioni nel prevedere la risposta del rivelatore, e la capacità di quest’ultimo nell’eseguire acquisizioni SPECT per ricostruzioni tomografiche con tempi compatibili con l’utilizzo clinico. Le prestazioni del rivelatore sono state infine simulate presso la struttura NUANS di Nagoya, in condizioni di irraggiamento con neutroni epitermici e con alto fondo. I risultati mostrano che è possibile ottenere proiezioni SPECT distinguibili entro tempi di acquisizione accettabili, fornendo una base per futuri miglioramenti e validazioni sperimentali.
Design optimization and Monte Carlo study of a SPECT imaging system for BNCT dose monitoring
Lazzarin, Emanuela
2024/2025
Abstract
Boron Neutron Capture Therapy (BNCT) is a radiotherapy technique that exploits the 10B(n,α)7Li reaction to kill tumor cells while sparing surrounding healthy tissues. An important aspect of this type of treatment is monitoring the dose delivery, that can be done in real time employing PG-SPECT. This work focuses on the development of BeNEdiCTE prototype detector, that aims at measuring and reconstructing the spatial distribution of the absorbed dose during BNCT treatments by the detection of the 478 keV photon, emitted in 94% of the neutron capture reactions on 10B. Monte Carlo simulations were performed on FLUKA to predict the performance of the detector and optimize its geometry. The spatial resolution of the collimator was evaluated, and subsequently improved by adding a lead patch. The optimized configuration was then implemented on FLUKA in the geometry representing the PGNAAline at LENA facility in Pavia, in order to simulate three different SPECT projection images of two borated vials, that were then used to obtain a 3D reconstruction through MLEM algorithm. The simulated projections were then compared to the results acquired during a measurement session carried out at LENA. The good agreement between simulations and experimental results demonstrates the ability of the simulations to predict the response of the detector, and its capability in performing SPECT acquisitions for tomographic reconstructions, with clinically compatible acquisition times. The performance of the detector was then simulated at the NUANS facility in Nagoya, under epithermal neutron irradiation field and high background condition. The results show that distinguishable SPECT projections can be obtained within acceptable acquisition times, providing a basis for further improvements and experimental validation.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_Lazzarin_Executive_Summary.pdf
non accessibile
Descrizione: Executive Summary
Dimensione
4.98 MB
Formato
Adobe PDF
|
4.98 MB | Adobe PDF | Visualizza/Apri |
|
2025_12_Lazzarin_Tesi.pdf
non accessibile
Descrizione: Tesi
Dimensione
20.2 MB
Formato
Adobe PDF
|
20.2 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246683