Type 1 diabetes is an autoimmune disease that impairs the ability to regulate blood glucose levels, causing chronic hyperglycemia and requiring continuous management with exogenous insulin. In this context, digital twins, dynamic virtual replicas of glucose-insulin metabolism, provide active support for complex therapeutic decisions through DSS systems. Well-established tools such as the UVa/Padova T1DS simulator and the ReplayBG framework allow in silico evaluation of insulin therapies but require more intuitive interfaces to facilitate their adoption in clinical and research settings. To meet these needs, the aim of this thesis is to develop a digital twin-based web tool for prototyping clinical decision support systems in Type 1 Diabetes management. The proposed system offers an interactive environment that allows users to simulate, modify and analyze the glycemic trend through a digital platform. The development of the web application is based on a modular architecture, with a frontend built in Flutter (Dart) and a backend built with Django and Python for data processing and integration with ReplayBG library. Communication between components takes place via RESTful API, ensuring flexibility and scalability. The application integrates CSV file upload functionality for customizing the model on patient data and encrypted packets containing pre-existing digital twins. The system supports the simulation of alternative scenarios through replays with modification of therapeutic parameters, providing AGP-based statistical analysis with calculation of glycemic metrics, as well as downloading results in various formats. The web platform developed is an operational tool to support research, development and training in type 1 diabetes management. Simulation technologies such as this also offer potential educational value, facilitating the understanding of treatment by patients and caregivers, and promoting conscious self-management of the disease.
Il diabete di tipo 1 è una patologia autoimmune che compromette la capacità di regolare i livelli di glucosio nel sangue, causando iperglicemia cronica e richiedendo una gestione continua con insulina esogena. In questo contesto, i digital twin, repliche virtuali dinamiche del metabolismo glucosio-insulinico, offrono un supporto attivo alle decisioni terapeutiche complesse attraverso sistemi DSS. Strumenti consolidati come il simulatore UVa/Padova T1DS e il framework ReplayBG permettono la valutazione in silico delle terapie insuliniche, ma richiedono interfacce più intuitive per favorirne l’adozione in ambito clinico e di ricerca. Per rispondere a queste esigenze, l’obiettivo di questa tesi è lo sviluppo di uno strumento web basato su digital twin per la prototipazione di sistemi di supporto alle decisioni cliniche nella gestione del Diabete di Tipo 1. Il sistema proposto offre un ambiente interattivo che consenta agli utenti di simulare, modificare e analizzare l’andamento glicemico attraverso una piattaforma digitale. Lo sviluppo dell'applicazione web si basa su un’architettura modulare, con un frontend realizzato in Flutter (Dart) e un backend costruito con Django e Python per l’elaborazione dei dati e l’integrazione con la libreria ReplayBG. La comunicazione tra componenti avviene tramite API RESTful, garantendo flessibilità e scalabilità. L'applicazione integra funzionalità di caricamento di file CSV per la personalizzazione del modello sui dati del paziente e di pacchetti cifrati contenenti digital twin preesistenti. Il sistema supporta la simulazione di scenari alternativi tramite replay con modifica di parametri terapeutici, fornendo analisi statistiche AGP-based con calcolo di metriche glicemiche, oltre al download di risultati in vari formati. La piattaforma web sviluppata rappresenta uno strumento operativo a supporto di ricerca, sviluppo e formazione nella gestione del diabete di tipo 1. Tecnologie di simulazione come questa offrono inoltre un potenziale valore educativo, facilitando la comprensione del trattamento da parte di pazienti e caregiver, e promuovendo l’autogestione consapevole della patologia.
Sviluppo di uno strumento web di digital-twin per il diabete di tipo 1 per la prototipazione di sistemi di supporto alla decisione
GHIOTTO, ANNA
2024/2025
Abstract
Type 1 diabetes is an autoimmune disease that impairs the ability to regulate blood glucose levels, causing chronic hyperglycemia and requiring continuous management with exogenous insulin. In this context, digital twins, dynamic virtual replicas of glucose-insulin metabolism, provide active support for complex therapeutic decisions through DSS systems. Well-established tools such as the UVa/Padova T1DS simulator and the ReplayBG framework allow in silico evaluation of insulin therapies but require more intuitive interfaces to facilitate their adoption in clinical and research settings. To meet these needs, the aim of this thesis is to develop a digital twin-based web tool for prototyping clinical decision support systems in Type 1 Diabetes management. The proposed system offers an interactive environment that allows users to simulate, modify and analyze the glycemic trend through a digital platform. The development of the web application is based on a modular architecture, with a frontend built in Flutter (Dart) and a backend built with Django and Python for data processing and integration with ReplayBG library. Communication between components takes place via RESTful API, ensuring flexibility and scalability. The application integrates CSV file upload functionality for customizing the model on patient data and encrypted packets containing pre-existing digital twins. The system supports the simulation of alternative scenarios through replays with modification of therapeutic parameters, providing AGP-based statistical analysis with calculation of glycemic metrics, as well as downloading results in various formats. The web platform developed is an operational tool to support research, development and training in type 1 diabetes management. Simulation technologies such as this also offer potential educational value, facilitating the understanding of treatment by patients and caregivers, and promoting conscious self-management of the disease.| File | Dimensione | Formato | |
|---|---|---|---|
|
Ghiotto_Anna.pdf
embargo fino al 22/10/2028
Dimensione
2.69 MB
Formato
Adobe PDF
|
2.69 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/95827