Human movement analysis is integral for diagnosing, managing, and treating orthopedic pathologies. Even if traditional marker-based Motion Capture (MoCap) systems have long been considered the gold standard for capturing kinematic data, they face several limitations due to costs and complexity of setup procedures. These challenges have driven interest in markerless systems, which leverage advancements in computer vision to provide a scalable and cost-effective alternative. Despite their potential, they require rigorous validation to ensure their accuracy and reliability for clinical use. This study investigates the use of clustering techniques to identify movement-based subgroups in patients while demonstrating a good level of agreement between marker-based and markerless systems for kinematic measurements, especially in the frontal and sagittal planes. In contrast, the reliability of the transverse plane was notably lower, highlighting an area for improvement in markerless systems. The clustering analysis was conducted using two methods: one based on relevant data points from the kinematic curves, and the other considering the whole time series. The results revealed distinct subgroups, underscoring the presence of biomechanical heterogeneity and providing valuable insights into movement patterns that could inform personalized rehabilitation strategies. To enhance the interpretability of the results and support the integration of data-driven methods into routine orthopedic assessment, a web-based application was developed, enabling interactive visualization of kinematic profiles. In conclusion, this study demonstrates the feasibility of integrating markerless MoCap systems with clustering techniques to complement traditional methods in orthopedic diagnostics. By providing a scalable and personalized approach to movement analysis, these systems, along with interactive visualization tools, have the potential to enhance clinical decision-making and improve patient outcomes.
L'analisi del movimento umano è fondamentale per la diagnosi, la gestione e il trattamento delle patologie ortopediche. Nonstante i sistemi di Motion Capture (MoCap) marker-based siano da tempo considerati il gold standard per l'acquisizione dei dati cinematici, presentano diverse limitazioni legate ai costi elevati e alla complessità delle procedure di configurazione. Queste criticità hanno alimentato l'interesse verso i sistemi markerless, che sfruttano i progressi della computer vision per offrire un'alternativa scalabile ed economicamente sostenibile. Nonostante il loro potenziale, tali sistemi richiedono una rigorosa validazione per garantirne l'accuratezza e l'affidabilità in ambito clinico. Questo studio indaga l'uso di tecniche di clustering per identificare sottogruppi basati sul movimento nei pazienti, dimostrando un buon livello di coerenza tra i sistemi marker-based e markerless nelle misurazioni cinematiche, in particolare nei piani frontale e sagittale. Al contrario, l'affidabilità nel piano trasversale è risultata significativamente inferiore, evidenziando un'area di miglioramento per i sistemi markerless. Il clustering è stato condotto utilizzando due metodologie: una basata su punti rilevanti delle curve cinematiche e l'altra considerando l'intera serie temporale. I risultati hanno evidenziato sottogruppi distinti, sottolineando la presenza di eterogeneità biomeccanica e fornendo indicazioni sui pattern di movimento, utili per orientare strategie riabilitative personalizzate. Per migliorare l'interpretabilità dei risultati e supportare l'integrazione di metodi data-driven nella pratica ortopedica, è stato sviluppata un'applicazione web che consente la visualizzazione dei profili cinematici. In conclusione, questo studio dimostra la fattibilità di integrare i sistemi MoCap markerless con tecniche di clustering per completare i metodi tradizionali nella diagnostica ortopedica. Offrendo un approccio scalabile e personalizzato all'analisi del movimento, questi sistemi, insieme agli strumenti di visualizzazione interattiva, hanno il potenziale per migliorare il processo decisionale clinico e gli esiti per i pazienti.
Movement-based clustering with integrated web visualization: bridging motion capture and clinical interpretation
Cabai, Eleonora
2024/2025
Abstract
Human movement analysis is integral for diagnosing, managing, and treating orthopedic pathologies. Even if traditional marker-based Motion Capture (MoCap) systems have long been considered the gold standard for capturing kinematic data, they face several limitations due to costs and complexity of setup procedures. These challenges have driven interest in markerless systems, which leverage advancements in computer vision to provide a scalable and cost-effective alternative. Despite their potential, they require rigorous validation to ensure their accuracy and reliability for clinical use. This study investigates the use of clustering techniques to identify movement-based subgroups in patients while demonstrating a good level of agreement between marker-based and markerless systems for kinematic measurements, especially in the frontal and sagittal planes. In contrast, the reliability of the transverse plane was notably lower, highlighting an area for improvement in markerless systems. The clustering analysis was conducted using two methods: one based on relevant data points from the kinematic curves, and the other considering the whole time series. The results revealed distinct subgroups, underscoring the presence of biomechanical heterogeneity and providing valuable insights into movement patterns that could inform personalized rehabilitation strategies. To enhance the interpretability of the results and support the integration of data-driven methods into routine orthopedic assessment, a web-based application was developed, enabling interactive visualization of kinematic profiles. In conclusion, this study demonstrates the feasibility of integrating markerless MoCap systems with clustering techniques to complement traditional methods in orthopedic diagnostics. By providing a scalable and personalized approach to movement analysis, these systems, along with interactive visualization tools, have the potential to enhance clinical decision-making and improve patient outcomes.File | Dimensione | Formato | |
---|---|---|---|
Executive_Summary_Cabai_Eleonora.pdf
accessibile in internet solo dagli utenti autorizzati
Descrizione: executive summary
Dimensione
780.36 kB
Formato
Adobe PDF
|
780.36 kB | Adobe PDF | Visualizza/Apri |
Thesis_Cabai_Eleonora.pdf
solo utenti autorizzati a partire dal 24/06/2026
Descrizione: thesis
Dimensione
7.29 MB
Formato
Adobe PDF
|
7.29 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/10589/239538