Operators involved in Interventional Radiology (IR) practices are subject to the highest occupational exposure among all categories of clinical personnel working with ionizing radiation, as they stand close to the patient while he/she is irradiated with the X-ray fluoroscopy machine. The legal dose assessment with personal dosimeters proved to be extremely inaccurate, given the highly heterogeneous field typical of IR setups. Moreover, it relies on the staff’s diligence in always using and correctly positioning the dosimeters. Computational Dosimetry can be used to overcome the typical limitations of personal dosimeters. Within the framework of the PODIUM research project (Personal Online DosImetry Using computational Methods), the Belgian Nuclear Research Center SCK CEN investigated the topic, developing computational tools for dose calculation in IR scenarios. In the present work, a new integrated computational system to track Interventional Radiology procedures and perform PHITS Monte Carlo simulations of radiation transport was developed, with the purpose of estimating the personal dose-equivalent Hp(10) to the operating staff. Inspired by the experience and main limitations of the PODIUM project, a brand-new tracking module was coded using state-of-the-art depth cameras with advanced body-tracking algorithms to locate the staff and ultra-wideband sensors to track a ceiling-suspended shielding device in the IR room. The system was installed in the OLV Hospital of Aalst, where it tracked several procedures. A data elaboration module, merging tracking and irradiation data and automatically generating sets of PHITS simulation input files, was also developed. The simulation results for six procedures, in terms of total Hp(10) for the group of operators involved, were then compared with the measurements from active personal dosimeters, showing a deviation between -59% and +40% of the corresponding measured Hp(10). Significant inaccuracies were observed for the ultra-wideband sensors used to track the ceiling-suspended shield. The use of a more accurate real-time location system or a secondary depth camera is suggested as an alternative strategy to track this shielding device.
Gli operatori coinvolti in pratiche di Radiologia Interventistica (RI) sono soggetti a dosi occupazionali estremamente elevate, dovendo operare in prossimità del paziente irradiato dal sistema di fluoroscopia a raggi X. I dosimetri personali indossati risultano altamente inaccurati, specialmente in relazione alla natura eterogenea del campo di radiazione tipico di tale pratica. L’uso di metodi di Dosimetria Computazionale come alternativa ai dosimetri può ovviare a questa limitazione. Il centro di ricerche nucleari SCK CEN, in Belgio, partecipando al progetto PODIUM (Personal Online DosImetry Using computational Methods), ha esplorato questo ambito sviluppando una serie di applicativi per il calcolo della dose associata a procedure di RI. Questa tesi magistrale presenta un nuovo sistema computazionale integrato per il monitoraggio di procedure di RI ed il calcolo della dose associata agli operatori coinvolti, tramite simulazioni Monte Carlo con il codice di trasporto di radiazione PHITS. Alla luce dell’esperienza e delle limitazioni del progetto PODIUM, un nuovo applicativo di tracking è stato sviluppato facendo uso di fotocamere di profondità di ultima generazione con algoritmi di body-tracking avanzati, abbinate a sensori a banda ultralarga per la localizzazione dello schermo radiologico utilizzato dal personale come dispositivo di protezione. Il sistema, installato nell’ospedale OLV di Aalst, è stato utilizzato durante diverse procedure di RI. Un modulo di elaborazione dati è stato inoltre sviluppato per creare automaticamente dei set di file di input PHITS a partire dai dati di tracking acquisiti e di irraggiamento contenuti nel software della macchina a raggi X. I risultati delle simulazioni effettuate per sei diverse procedure sono stati ottenuti in termini dell’equivalente di dose personale Hp(10) per il gruppo di operatori coinvolti in ciascuna pratica, e mostrano una deviazione contenuta tra -59% e +40% dai rispettivi valori misurati tramite dosimetri attivi. Inaccuratezze osservate nel sistema di localizzazione di sensori a banda ultralarga suggeriscono l’uso di sistemi più accurati o di una telecamera secondaria per il tracking dello schermo radiologico.
Development and initial validation of an integrated computational system for worker dosimetry in interventional radiology
Martelengo, Alessandro
2024/2025
Abstract
Operators involved in Interventional Radiology (IR) practices are subject to the highest occupational exposure among all categories of clinical personnel working with ionizing radiation, as they stand close to the patient while he/she is irradiated with the X-ray fluoroscopy machine. The legal dose assessment with personal dosimeters proved to be extremely inaccurate, given the highly heterogeneous field typical of IR setups. Moreover, it relies on the staff’s diligence in always using and correctly positioning the dosimeters. Computational Dosimetry can be used to overcome the typical limitations of personal dosimeters. Within the framework of the PODIUM research project (Personal Online DosImetry Using computational Methods), the Belgian Nuclear Research Center SCK CEN investigated the topic, developing computational tools for dose calculation in IR scenarios. In the present work, a new integrated computational system to track Interventional Radiology procedures and perform PHITS Monte Carlo simulations of radiation transport was developed, with the purpose of estimating the personal dose-equivalent Hp(10) to the operating staff. Inspired by the experience and main limitations of the PODIUM project, a brand-new tracking module was coded using state-of-the-art depth cameras with advanced body-tracking algorithms to locate the staff and ultra-wideband sensors to track a ceiling-suspended shielding device in the IR room. The system was installed in the OLV Hospital of Aalst, where it tracked several procedures. A data elaboration module, merging tracking and irradiation data and automatically generating sets of PHITS simulation input files, was also developed. The simulation results for six procedures, in terms of total Hp(10) for the group of operators involved, were then compared with the measurements from active personal dosimeters, showing a deviation between -59% and +40% of the corresponding measured Hp(10). Significant inaccuracies were observed for the ultra-wideband sensors used to track the ceiling-suspended shield. The use of a more accurate real-time location system or a secondary depth camera is suggested as an alternative strategy to track this shielding device.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_Martelengo_Tesi.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Elaborato di tesi
Dimensione
10.4 MB
Formato
Adobe PDF
|
10.4 MB | Adobe PDF | Visualizza/Apri |
|
2025_10_Martelengo_Executive Summary.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: Executive summary
Dimensione
4.44 MB
Formato
Adobe PDF
|
4.44 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/10589/243023