This thesis presents the design, development, and validation of a low-cost telerehabilitation system based on wearable inertial sensors (IMUs) and an IoT infrastructure for remotely monitoring motor and respiratory exercises. The project addresses the growing need for accessible and personalized rehabilitation solutions, particularly for individuals with chronic conditions. The system integrates prototypical sensors, a mobile application, a rehabilitation protocol targeting the upper and lower limbs, trunk, and respiratory muscles, as well as key performance indicators related to various motor functions. Special emphasis was placed on usability, data privacy, and affordability, aligning with the United Nations 2030 Sustainable Development Goals (SDGs 3, 10, and 12). A series of validation studies were conducted in laboratory, clinical, and home settings. The results confirm the system’s accuracy in measuring repetitions, execution time, movement intensity, and smoothness; its validity in detecting movement errors through machine learning techniques; and its potential for clinical application, for example in chronic low back pain. A parallel study also validated the SISTINE system, an extension of the platform incorporating sensorized socks, for balance and gait evaluation. Despite some current limitations in estimating specific indicators, the system proved to be reliable for low-intensity rehabilitation and suitable for unsupervised home use. Future developments include the integration of TinyML for real-time feedback, hardware improvements, and user interface enhancements, with the goal of achieving CE certification and enabling wider adoption. This work demonstrates that technological innovation, when combined with human movement science, can significantly
La presente tesi descrive la progettazione, lo sviluppo e la validazione di un sistema di teleriabilitazione a basso costo basato su sensori inerziali indossabili (IMU) e un’infrastruttura IoT per il monitoraggio da remoto di esercizi motori e respiratori. Il progetto risponde al crescente bisogno di soluzioni riabilitative accessibili e personalizzate, in particolare per persone affette da patologie croniche. Il sistema integra sensori prototipali, un’applicazione mobile, un protocollo riabilitativo mirato agli arti superiori, inferiori, tronco e muscoli respiratori, e indicatori di performance relativi a diverse capacità motorie. Particolare attenzione è stata posta su usabilità, protezione dei dati e accessibilità economica, in linea con gli Obiettivi di Sviluppo Sostenibile delle Nazioni Unite 2030 (SDGs 3, 10 e 12). Sono stati condotti studi di validazione in laboratorio, in contesti clinici e domiciliari. I risultati confermano l’accuratezza del sistema nel misurare ripetizioni, tempi di esecuzione, intensità e fluidità del movimento; la sua validità nel rilevare errori motori tramite tecniche di machine learning; e il suo potenziale per applicazioni cliniche, in particolare nella gestione della lombalgia cronica. Uno studio parallelo ha inoltre validato il sistema SISTINE, un’estensione dell’infrastruttura con calze sensorizzate, per la valutazione dell’equilibrio e del cammino. Nonostante alcune limitazioni nella stima di specifici indicatori, il sistema si è dimostrato affidabile per la riabilitazione a bassa intensità e utilizzabile in autonomia a domicilio. Gli sviluppi futuri prevedono l’integrazione del TinyML per un feedback in tempo reale, il miglioramento hardware e l’ottimizzazione dell’interfaccia utente, con l’obiettivo di ottenere la certificazione CE e una diffusione su larga scala. Questo lavoro dimostra come l’innovazione tecnologica, integrata con le scienze del movimento umano, possa migliorare in modo significativo la riabilitazione a distanza, rendendola più efficace, personalizzata e accessibile.
Development of a telerehabilitation care pathway based on respiratory and motor reactivation exercises for patients with chronic respiratory diseases or as an outcome of COVID 19
CARAMIA, FEDERICO
2025
Abstract
This thesis presents the design, development, and validation of a low-cost telerehabilitation system based on wearable inertial sensors (IMUs) and an IoT infrastructure for remotely monitoring motor and respiratory exercises. The project addresses the growing need for accessible and personalized rehabilitation solutions, particularly for individuals with chronic conditions. The system integrates prototypical sensors, a mobile application, a rehabilitation protocol targeting the upper and lower limbs, trunk, and respiratory muscles, as well as key performance indicators related to various motor functions. Special emphasis was placed on usability, data privacy, and affordability, aligning with the United Nations 2030 Sustainable Development Goals (SDGs 3, 10, and 12). A series of validation studies were conducted in laboratory, clinical, and home settings. The results confirm the system’s accuracy in measuring repetitions, execution time, movement intensity, and smoothness; its validity in detecting movement errors through machine learning techniques; and its potential for clinical application, for example in chronic low back pain. A parallel study also validated the SISTINE system, an extension of the platform incorporating sensorized socks, for balance and gait evaluation. Despite some current limitations in estimating specific indicators, the system proved to be reliable for low-intensity rehabilitation and suitable for unsupervised home use. Future developments include the integration of TinyML for real-time feedback, hardware improvements, and user interface enhancements, with the goal of achieving CE certification and enabling wider adoption. This work demonstrates that technological innovation, when combined with human movement science, can significantlyFile | Dimensione | Formato | |
---|---|---|---|
PhD_Thesis_Final_Caramia_Federico.pdf
accesso aperto
Dimensione
19.4 MB
Formato
Adobe PDF
|
19.4 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/20.500.14242/212342
URN:NBN:IT:UNIROMA4-212342