<< Torna alla home page
L'inserimento della tesi può essere fatto dall'autore stesso, da un responsabile d'esperimento o dalle segreterie scientifiche
Per qualsiasi problema di natura tecnica scrivere a Supporto Web INFN

NEW: E' assolutamente vietato inserire il nome del relatore al posto del tesista per "bypassare" l'inserimento in anagrafica. Le tesi così formate saranno ritenute invalide ed eliminate dal sistema

Il tesista dev'essere presente nel database dell'anagrafica centralizzata (GODiVA)
Cercare il tesista per cognome, cliccando sul pulsante apposito, ed eventualmente inserirlo tramite il link proposto. Sono richieste le informazioni fondamentali di anagrafica e tipo/numero di documento di riconoscimento
Ultimo aggiornamento 22 apr 2019
Autore
Alberto Taffelli
Sesso M
Esperimento MOVE_IT
Tipo Laurea Magistrale
Destinazione dopo il cons. del titolo Dottorato (Italia)
Università Universita' Di Trento
Strutt.INFN/Ente
TIFPA
Titolo Modeling the risk of toxicity in brain tumor patients treated with protons
Abstract Radiation induced toxicity events may a ect the quality of life (QoL) of patients after a radiotherapy treatment. Our aim was rst to develop an NTCP model, based on the treatment dose maps, for radiation-induced fa- tigue (RIF) and radiation-induced alopecia (RIA), in patients treated with scanning beam proton therapy (PT) for brain tumors (BT). Furthermore, we are interested in assessing the impact on the derived models when applying a variable model of RBE on dose maps. We evaluated 85 BT patients undergoing PT by pencil beam scanning in a retroprospective analysis assessing acute (< 90 days) and late (> 90 days) RIF and RIA, classi ed according to the CTCAE v.4 scoring system. Brain- stem and scalp were considered the targets of radiation for RIF and RIA, respectively. DVHs of the brainstem and DSHs of the scalp were extracted together with Dmax and Dmean. Patient and treatment-related characteris- tics were analyzed with dose metrics extracted. Chi-square/Mann-Whitney tests were employed for univariate statistical analysis. NTCP models by multivariate logistic regression were developed and model performance was measured by the AUC ROC. Validation of the RBE calculation available in the TPS of the facility was performed using a Monte Carlo simulation by Geant4, prior to apply it to a subset of the initial patient database. RIF analysis does not show any statistically signi cant association between the variables considered (either clinical or dosimetric) and the outcome. RIF analysis, on the contrary, reveals strongly signi cant associations between extracted variables and outcome. NTCP models derived for acute and late RIA show very good prediction performance (AUC > 0.8) in determining the onset of toxicity. Moreover, the application of a variable RBE model on the patient subsets shows statistically signi cant di erences in the dose metrics between the original plans and the recalculated one and hence, the need to extend the analysis to the whole database in order to assess variations in the NTCP models.
Anno iscrizione 2017
Data conseguimento 18 lug 2018
Luogo conseguimento Trento
Relatore/i
Francesco Tommasino Emanuele Scifoni  
File PDF
TaffelliMasterthesis.pdf
File PS