The healthcare system is undergoing changes: continuous and real-time monitoring, gath ering of an increasing amount of data, and early disease detection represent factors that decentralize the hospital, moving closer to personalized medicine and improving patients’ health and life quality. The objective of this Thesis is the design of a smart ring capable of monitoring pulse rate and oxygen saturation. A ring-shaped pulse oximeter provides both the convenience of wrist-mounted pulse oximeters and the accuracy of fingertip pulse oximeters, as the measurement location is closer to the fingertip, where there is greater perfusion compared to the wrist. Therefore, building upon previous work, which involved the development of a wrist worn pulse oximeter, the aim is to relocate it to the base of the finger. As it is a wearable device, the data is sent through the ANT protocol to a mobile appli cation that can be installed on a smartphone. Through this application, it is possible to observe the data in real-time. After the development of a prototype, made to assess the feasibility of acquiring accurate data from the new measurement site, a rigid-flex PCB was developed to take the shape of a ring. The validation phase differed for the prototype and the final device. In fact, the smart ring was tested not only during the repetition of apnea periods but also during cycling activities to monitor an increase in heart rate. The Bland – Altman (BA) analysis was implemented to validate the ring against the gold standard (Nonin’s WristOx2), demonstrating the validity of the data. During device testing, the Bland-Altman analysis for subject 1 revealed a bias of 0.49 and limits of agreement (LoA) of [15.38, -14.40], for pulse rate values against the gold standard. Similarly, for oxygen saturation, the bias was 0.99, and the LoA were [3.09, -1.10]. This indicates a systematic slight overestimation error for SpO2, while maintaining physiologically acceptable limits of agreement for both measurements.
Il sistema sanitario sta attraversando un cambiamento: il monitoraggio continuo e in tempo reale, il raccoglimento di sempre più dati, il rilevamento prematuro di una malattia sono fattori che decentralizzano l’ospedale, avvicinandosi alla medicina personalizzata e al miglioramento della salute e della qualità di vita dei pazienti. In questo contesto, i dispositivi indossabili giocano un ruolo fondamentale. L’obiettivo di questa Tesi consiste nella progettazione di uno smart ring, per monitorare la frequenza cardiaca e la saturazione. Un pulsossimetro ad anello permette di raggiungere sia la comodità fornita da quelli da polso che l’accuratezza di quelli da dito, dove si trova una maggiore irrorazione. Quindi, partendo dal lavoro precedente, in cui è stato sviluppato un pulsossimetro da polso, si vuole spostarlo alla base del dito. Essendo un dispositivo indossabile, i dati vengono inviati via protocollo ANT ad una applicazione per cellulare, attraverso la quale è possibile osservare i dati in tempo reale. Successivamente allo sviluppo di un prototipo, per valutare la fattibilità dell’acquisizione di dati veritieri dal nuovo sito di misurazione, una PCB rigida-flessibile, per creare la forma di un anello, è stata sviluppata. La validazione è stata differente per prototipo e dispositivo finale. Infatti, lo smart ring è stato testato non solo durante la ripetizione di periodi di apnea, ma anche durante un’attività su cyclette, in modo da monitorare un incremento del battito. L’analisi di Bland – Altman è stata implementata rispetto al gold standard (WristOx2 di Nonin), dimostrando la validità dei dati. Durante il test del dispositivo, l’analisi di Bland-Altman per il soggetto 1 ha rivelato un bias di 0.49 e limiti di accordo (LoA) di [15.38, -14.40], per i valori della frequenza cardiaca rispetto allo standard di riferimento. Allo stesso modo, per la saturazione, il bias è stato 0.99 e i LoA sono stati [3.09, -1.10]. Ciò indica un leggero errore sistematico di sovrastima per la SpO2, pur mantenendo LoA fisiologicamente accettabili per entrambe le misurazioni.
Design, development and testing of a smart ring to monitor pulse rate and oxygen saturation
Montenegro, Martina
2022/2023
Abstract
The healthcare system is undergoing changes: continuous and real-time monitoring, gath ering of an increasing amount of data, and early disease detection represent factors that decentralize the hospital, moving closer to personalized medicine and improving patients’ health and life quality. The objective of this Thesis is the design of a smart ring capable of monitoring pulse rate and oxygen saturation. A ring-shaped pulse oximeter provides both the convenience of wrist-mounted pulse oximeters and the accuracy of fingertip pulse oximeters, as the measurement location is closer to the fingertip, where there is greater perfusion compared to the wrist. Therefore, building upon previous work, which involved the development of a wrist worn pulse oximeter, the aim is to relocate it to the base of the finger. As it is a wearable device, the data is sent through the ANT protocol to a mobile appli cation that can be installed on a smartphone. Through this application, it is possible to observe the data in real-time. After the development of a prototype, made to assess the feasibility of acquiring accurate data from the new measurement site, a rigid-flex PCB was developed to take the shape of a ring. The validation phase differed for the prototype and the final device. In fact, the smart ring was tested not only during the repetition of apnea periods but also during cycling activities to monitor an increase in heart rate. The Bland – Altman (BA) analysis was implemented to validate the ring against the gold standard (Nonin’s WristOx2), demonstrating the validity of the data. During device testing, the Bland-Altman analysis for subject 1 revealed a bias of 0.49 and limits of agreement (LoA) of [15.38, -14.40], for pulse rate values against the gold standard. Similarly, for oxygen saturation, the bias was 0.99, and the LoA were [3.09, -1.10]. This indicates a systematic slight overestimation error for SpO2, while maintaining physiologically acceptable limits of agreement for both measurements.File | Dimensione | Formato | |
---|---|---|---|
2024_04_Montenegro.pdf
accessibile in internet per tutti
Dimensione
13.96 MB
Formato
Adobe PDF
|
13.96 MB | Adobe PDF | Visualizza/Apri |
2024_04_Montenegro_Executive_Summary.pdf
accessibile in internet per tutti
Dimensione
2.19 MB
Formato
Adobe PDF
|
2.19 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/10589/219411