This thesis focuses on the application and critical evaluation of weighting methodologies, with particular attention to the comparison between two data sources with Individual Patient Data (IPD). The main objective of the research was to investigate inference techniques in the context of weighted data through simulations based on a real study. I was asked by a company specializing in neuromodulation and cardiovascular surgery medical devices to design a prospective, multicenter, observational study with parallel groups to evaluate the efficacy and safety of Vagus Nerve Stimulation (VNS) therapy compared to the Standard of Care pharmacological treatment. VNS is an implantable device for the treatment of epilepsy. The primary endpoint of the study was the maximum duration of the treatment response, quantified in months. The thesis provides an overview of the weighting method used and the different approaches to inference for weighted data, assessing their potential, limitations and appropriateness.
Il presente lavoro di tesi si concentra sull'applicazione e la valutazione critica di metodologie di weighting, con particolare attenzione al confronto tra due fonti di dati con Individual Patient Data (IPD). L'obiettivo principale della ricerca è stato indagare le tecniche di inferenza nel contesto di dati pesati, attraverso simulazioni basate su uno studio reale. Mi è stato chiesto da una società che si occupa di dispositivi medici per neuromodulazione e chirurgia cardiovascolare di diegnare uno studio prospettico, multicentrico, osservazionale, a gruppi paralleli, per valutare l’efficacia e la sicurezza della terapia Vagus Nerve Stimulation (VNS) confrontata con lo Standard of Care farmacologico. Il VNS è un device impiantabile per il trattamento dell’epilessia. L’endpoint primario dello studio era la durata massima della risposta al trattamento, quantificata in mensilità. Nella tesi si offre una panoramica del metodo di weighting utilizzato e sui diversi approcci all'inferenza per i dati pesati, valutandone criticità, potenzialità e adeguatezza.
INFERENZA SU DATI PESATI: APPLICAZIONE E SIMULAZIONI IN UN CASO DI STUDIO
PARODI, ANDREA
2025
Abstract
This thesis focuses on the application and critical evaluation of weighting methodologies, with particular attention to the comparison between two data sources with Individual Patient Data (IPD). The main objective of the research was to investigate inference techniques in the context of weighted data through simulations based on a real study. I was asked by a company specializing in neuromodulation and cardiovascular surgery medical devices to design a prospective, multicenter, observational study with parallel groups to evaluate the efficacy and safety of Vagus Nerve Stimulation (VNS) therapy compared to the Standard of Care pharmacological treatment. VNS is an implantable device for the treatment of epilepsy. The primary endpoint of the study was the maximum duration of the treatment response, quantified in months. The thesis provides an overview of the weighting method used and the different approaches to inference for weighted data, assessing their potential, limitations and appropriateness.File | Dimensione | Formato | |
---|---|---|---|
phd_unimib_079343.pdf
accesso aperto
Dimensione
1.08 MB
Formato
Adobe PDF
|
1.08 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/298017
URN:NBN:IT:UNIMIB-298017