This thesis represents the second phase of a project that began in a previous thesis work (Corsini et al.), where the OpenSense ToolBox was specifically modified to be applied to data from the Cometa acquisition system. The project was carried out within the Movement Bioengineering Laboratory at the University of Padua. A group of seven healthy subjects was involved, in addition to the four previously studied, in order to increase the robustness of the comparison and validate the musculoskeletal model during gait. The project focuses on joint kinematics, that is, the study of the relative movement between two adjacent bone segments, with particular attention to the lower limb joints during walking. Two methods were employed to capture movement: one based on inertial sensors (IMU) and the other on stereophotogrammetry. In the analysis conducted, the primary objective was to compare the data collected from a subject using two different types of inertial sensors: XSens sensors, processed via the IK IMU-Driven ToolBox implemented in OpenSim, specifically designed for this device, and Cometa sensors. The goal of this phase is to validate the use of Cometa sensors in OpenSim using the same ToolBox, to verify their accuracy and reliability compared to the already established solution with XSens sensors. Once this validation is confirmed, the next phase will involve directly com paring the data acquired with Cometa sensors with those obtained via stereophotogrammetry, recognized as the gold standard for motion analysis. Both the raw data from the two different acquisition systems and the data processed with the OpenSim toolbox from both methods were compared. Finally, the performance of the model developed in this study was compared with the data from Al Borno et al., to assess the effectiveness of using Cometa IMUs, processed through the IMU-driven Toolbox in OpenSim, in calculating lower limb joint kinematics. The comparison between the various motion capture approaches was performed through graphical representations of joint angles as a function of the gait cycle percentage and the analysis of various statistical parameters. This comparison allowed for model validation, highlighting both the advantages of the sensors used and the areas requiring further future studies.
Il seguente lavoro di tesi rappresenta la seconda fase di un progetto iniziato in un precedente lavoro di tesi (Corsini et al.), nel quale la ToolBox OpenSense è stata appositamente modificata per essere applicata su dati provenienti da sistema di acquisizione Cometa. Il progetto si è svolto all’interno del laboratorio di Bioingegneria del Movimento presso l’Università di Padova. È stato coinvolto un gruppo di 7 soggetti sani, in aggiunta ai 4 precedentemente studiati, così da aumentare la robustezza della comparazione e validare il modello muscolo-scheletrico durante la camminata. Il progetto si concentra sulla cinematica articolare, ovvero lo studio del movimento relativo tra due segmenti ossei adiacenti, con particolare attenzione alle articolazioni degli arti inferiori durante il cammino. Sono stati impiegati due metodi per catturare il movimento: uno basato su sensori inerziali (IMU) e l’altro sulla stereofotogrammetria. Nell’analisi condotta, il primo obiettivo è stato confrontare i dati raccolti su un soggetto utilizzando due diverse tipologie di sensori inerziali: i sensori XSens, elaborati tramite la ToolBox IK IMU-Driven implementata su OpenSim, progettata appositamente per questo dispositivo, e i sensori Cometa. Lo scopo di questa fase è validare l’uso dei sensori Cometa su OpenSim tramite la medesima ToolBox, al fine di verificarne l’accuratezza e l’affidabilità rispetto alla soluzione già consolidata con i sensori XSens. Una volta che questa validazione sarà confermata, si procederà alla fase successiva, nella quale i dati acquisiti con i sensori Cometa saranno confrontati direttamente con quelli ottenuti mediante stereofotogrammetria, riconosciuta come gold standard per l’analisi del movimento. Infine, le prestazioni del modello sviluppato in questo studio sono state confrontate con i dati di Al Borno et al., per verificare l’attendibilità delle variabili stimate tramite degli IMU Cometa, elaborati attraverso la Toolbox IMU-driven di OpenSim, nel calcolo della cinematica articolare degli arti inferiori. Il confronto tra i vari approcci di rilevamento del movimento è stato eseguito tramite rappresentazioni grafiche degli angoli in funzione della percentuale del ciclo del passo e l’analisi di vari parametri statistici. Questo confronto ha consentito di validare il modello, mettendo in evidenza sia i vantaggi dei sensori utilizzati sia i limiti che richiedono ulteriori studi futuri.
Sviluppo di Modelli Muscolo-Scheletrici a partire da Sensori Inerziali in OpenSim: verifica di accuratezza e ripetibilità
LAZZARINI, ASIA
2023/2024
Abstract
This thesis represents the second phase of a project that began in a previous thesis work (Corsini et al.), where the OpenSense ToolBox was specifically modified to be applied to data from the Cometa acquisition system. The project was carried out within the Movement Bioengineering Laboratory at the University of Padua. A group of seven healthy subjects was involved, in addition to the four previously studied, in order to increase the robustness of the comparison and validate the musculoskeletal model during gait. The project focuses on joint kinematics, that is, the study of the relative movement between two adjacent bone segments, with particular attention to the lower limb joints during walking. Two methods were employed to capture movement: one based on inertial sensors (IMU) and the other on stereophotogrammetry. In the analysis conducted, the primary objective was to compare the data collected from a subject using two different types of inertial sensors: XSens sensors, processed via the IK IMU-Driven ToolBox implemented in OpenSim, specifically designed for this device, and Cometa sensors. The goal of this phase is to validate the use of Cometa sensors in OpenSim using the same ToolBox, to verify their accuracy and reliability compared to the already established solution with XSens sensors. Once this validation is confirmed, the next phase will involve directly com paring the data acquired with Cometa sensors with those obtained via stereophotogrammetry, recognized as the gold standard for motion analysis. Both the raw data from the two different acquisition systems and the data processed with the OpenSim toolbox from both methods were compared. Finally, the performance of the model developed in this study was compared with the data from Al Borno et al., to assess the effectiveness of using Cometa IMUs, processed through the IMU-driven Toolbox in OpenSim, in calculating lower limb joint kinematics. The comparison between the various motion capture approaches was performed through graphical representations of joint angles as a function of the gait cycle percentage and the analysis of various statistical parameters. This comparison allowed for model validation, highlighting both the advantages of the sensors used and the areas requiring further future studies.File | Dimensione | Formato | |
---|---|---|---|
Lazzarini_Asia.pdf
accesso riservato
Dimensione
14.04 MB
Formato
Adobe PDF
|
14.04 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/20.500.12608/73142