Glucose clamps are experimental techniques used to assets insulin action and glucose utilization. By ”clamping” the blood glucose of the patient to a desired level, insulin sensitivity and β-cell functioning can be investigated. In this work we present an approach to modulate glucose infusion rate (GIR) through a semiautomated clamp technique, serving as a decision support system to clinicians. We make two key contributions. Firstly, we investigate the application of control algorithms, specifically Proportional-Integral-Derivative (PID) controllers and Model Predictive Control (MPC), in the context of glucose clamp experiments. Using MATLAB and Simulink simulations, the two approaches were evaluated in both ideal and realistic scenarios, considering measurement variations and uncertainties typical in clinical environments. The results obtained were promising, demonstrating the efficacy of our semi-automated clamp technique in achieving targeted blood glucose (BG) levels. Both the ProportionalIntegral-Derivative (PID) control algorithm and the Model Predictive Control (MPC) approach effectively modulated the glucose infusion rate (GIR), leading to close adherence to predefined clamp BG targets. The second contribution is the design of a mobile application, Glucas 2.0, to support clinical researchers conducting glucose clamps experiments. This app, which enables the possibility to carry out more than one experiment at the time, provides suggestion of glucose infusion rates (GIR) generated by a PID controller.
I clamp glicemici sono una tecnica sperimentale utilizzata per valutare l'azione dell'insulina e l'utilizzazione del glucosio. Fissando il glucosio nel sangue del paziente a un livello desiderato, è possibile investigare la sensibilità all'insulina e il funzionamento delle cellule β. In questo lavoro presentiamo un approccio per modulare il tasso di infusione del glucosio (GIR) attraverso una tecnica di clamp semiautomatizzata, che funge da sistema di supporto decisionale per i clinici. Apportiamo due contributi chiave. In primo luogo, indaghiamo sull'applicazione di algoritmi di controllo, in particolare controller Proporzionale-Integrale-Derivativo (PID) e Controllo Predittivo del Modello (MPC), nel contesto degli esperimenti di clamp glicemici. Utilizzando simulazioni MATLAB e Simulink, i due approcci sono stati valutati sia in scenari ideali che realistici, considerando variazioni delle misurazioni e incertezze tipiche degli ambienti clinici. I risultati ottenuti sono stati promettenti, dimostrando l'efficacia della nostra tecnica di clamp semiautomatizzata nel raggiungere i livelli di glucosio nel sangue (BG) target. Sia l'algoritmo di controllo Proporzionale-Integrale-Derivativo (PID) che l'approccio del Controllo Predittivo del Modello (MPC) hanno modulato efficacemente il tasso di infusione del glucosio (GIR), portando a un'adesione stretta agli obiettivi predefiniti del clamp BG. Il secondo contributo è il design di un'applicazione mobile, Glucas 2.0, per supportare i ricercatori clinici che conducono esperimenti di clamp del glucosio. Questa app, che consente la possibilità di eseguire più di un esperimento contemporaneamente, fornisce suggerimenti sui tassi di infusione del glucosio (GIR) generati da un controller PID.
Design of closed-loop algorithms and of a decision support app for glucose clamp experiments
TUBARO, GIOVANNI
2023/2024
Abstract
Glucose clamps are experimental techniques used to assets insulin action and glucose utilization. By ”clamping” the blood glucose of the patient to a desired level, insulin sensitivity and β-cell functioning can be investigated. In this work we present an approach to modulate glucose infusion rate (GIR) through a semiautomated clamp technique, serving as a decision support system to clinicians. We make two key contributions. Firstly, we investigate the application of control algorithms, specifically Proportional-Integral-Derivative (PID) controllers and Model Predictive Control (MPC), in the context of glucose clamp experiments. Using MATLAB and Simulink simulations, the two approaches were evaluated in both ideal and realistic scenarios, considering measurement variations and uncertainties typical in clinical environments. The results obtained were promising, demonstrating the efficacy of our semi-automated clamp technique in achieving targeted blood glucose (BG) levels. Both the ProportionalIntegral-Derivative (PID) control algorithm and the Model Predictive Control (MPC) approach effectively modulated the glucose infusion rate (GIR), leading to close adherence to predefined clamp BG targets. The second contribution is the design of a mobile application, Glucas 2.0, to support clinical researchers conducting glucose clamps experiments. This app, which enables the possibility to carry out more than one experiment at the time, provides suggestion of glucose infusion rates (GIR) generated by a PID controller.File | Dimensione | Formato | |
---|---|---|---|
Tubaro_Giovanni.pdf
accesso aperto
Dimensione
11.7 MB
Formato
Adobe PDF
|
11.7 MB | Adobe PDF | Visualizza/Apri |
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/20.500.12608/64492