The Ponseti method stands as the primary non-surgical intervention for congenital clubfoot, a prevalent pediatric musculoskeletal malformation characterized by adduction, cavus, varus, and equinus deformities. The primary goal of this treatment is to correct deformities through a regimen of weekly manipulations and casting. Subsequently, an abduction orthosis is applied to not only sustain the achieved correction but also to forestall any potential relapses. While demanding substantial commitment from families and hospitals with weekly visits, the success rate of this treatment is high; however, there exists a relapse possibility ranging from 10% to 30%. The initiative to develop an orthosis capable of replicating Ponseti method movements and ensuring lasting correction is currently underway at the Department of ICT and Science at the University of Aalesund, Norway. To achieve this goal, it is imperative to gain a profound understanding and precise quantification of the corrective steps of the Ponseti method. Leveraging a motion capture system supported by ArUco markers, this study aims to track the dynamic movements of the foot during the treatment, generating a visualization of the maneuvers performed by the clinician. The collaboration with Aalesund Hospital and orthopedic specialists has played a pivotal role in shaping the experimental setup and the subsequent phases of data analysis. Twelve strategically placed markers on an articulated skeletal foot model were captured by three cameras while the clinician performed the Ponseti manipulation, providing posterior, lateral, and medial views. Crucial points of interest include the alignment of the calcaneus with the talus and the navicular bone with the talus head, essential for proper foot correction. The dynamics of these markers were then extrapolated. Despite the constraints of a limited dataset, commendable repeatability in the results was attained, ensuring a certain reliability of measurements and robustness in the analysis. However, the intrinsic complexity of the Ponseti method, combined with potential variations in its execution, highlights the necessity to conduct a more extensive range of tests, considering various cases involving different surgeons. This approach will facilitate the development of a more comprehensive dataset, allowing for a more precise generalization of the Ponseti method across a wider spectrum of clinical scenarios.
Il metodo Ponseti si configura come il principale intervento non chirurgico per il piede torto congenito, una diffusa malformazione muscolo-scheletrica pediatrica caratterizzata da adduzione, cavismo, varismo ed equinismo. L’obiettivo primario di questo trattamento è la completa correzione delle deformità attraverso manipolazioni settimanali e gessature, seguite dall’applicazione di un’ortesi di abduzione per sostenere la correzione ottenuta e prevenire ricadute. Pur richiedendo un notevole impegno da parte delle famiglie e dell’ospedale con visite settimanali, la percentuale di successo di questo trattamento è elevata, tuttavia si registra una possibilità di ricaduta compresa tra il 10% e il 30%. L’idea di sviluppare un’ortesi in grado di replicare i movimenti previsti dal metodo Ponseti e garantire una correzione duratura è un’iniziativa in corso presso il Dipartimento di ICT e Scienze dell’Università di Aalesund, in Norvegia. Per conseguire questo obiettivo, è essenziale comprendere a fondo e quantificare con precisione le fasi correttive del metodo Ponseti. Sfruttando un sistema di motion capture supportato da marcatori ArUco, questo studio mira a tracciare i movimenti dinamici del piede durante il trattamento, generando una visualizzazione delle manovre svolte dal clinico. La collaborazione con l’Ospedale di Aalesund e con gli ortopedici ha rivestito un ruolo fondamentale nella definizione del setup sperimentale e nelle successive fasi di analisi dei dati. Dodici marcatori, posizionati strategicamente su un modello scheletrico articolato di piede, sono stati ripresi da tre telecamere durante l’esecuzione della manipolazione da parte del clinico, fornendo una visione posteriore, laterale e mediale. I punti cruciali di interesse riguardano l’allineamento del calcaneo con l’astragalo e del navicolare con la testa dell’astragalo, essenziali per una corretta correzione del piede. La dinamica di questi marcatori è stata poi estrapolata. Nonostante i limiti di un dataset vincolato, è stata raggiunta un’apprezzabile ripetibilità dei risultati, garantendo una certa affidabilità delle misurazioni e solidità dell’analisi. Tuttavia, la complessità intrinseca del metodo Ponseti, unita alle potenziali variazioni nella sua esecuzione, sottolinea la necessità di condurre una gamma più ampia di test coinvolgendo diversi chirurghi. Questo approccio faciliterà lo sviluppo di un dataset più completo, consentendo una generalizzazione più accurata del metodo Ponseti per un più ampio spettro di scenari clinici.
Estimating axis alignment during Ponseti method on clubfoot using computer vision
Miele, Chiara
2022/2023
Abstract
The Ponseti method stands as the primary non-surgical intervention for congenital clubfoot, a prevalent pediatric musculoskeletal malformation characterized by adduction, cavus, varus, and equinus deformities. The primary goal of this treatment is to correct deformities through a regimen of weekly manipulations and casting. Subsequently, an abduction orthosis is applied to not only sustain the achieved correction but also to forestall any potential relapses. While demanding substantial commitment from families and hospitals with weekly visits, the success rate of this treatment is high; however, there exists a relapse possibility ranging from 10% to 30%. The initiative to develop an orthosis capable of replicating Ponseti method movements and ensuring lasting correction is currently underway at the Department of ICT and Science at the University of Aalesund, Norway. To achieve this goal, it is imperative to gain a profound understanding and precise quantification of the corrective steps of the Ponseti method. Leveraging a motion capture system supported by ArUco markers, this study aims to track the dynamic movements of the foot during the treatment, generating a visualization of the maneuvers performed by the clinician. The collaboration with Aalesund Hospital and orthopedic specialists has played a pivotal role in shaping the experimental setup and the subsequent phases of data analysis. Twelve strategically placed markers on an articulated skeletal foot model were captured by three cameras while the clinician performed the Ponseti manipulation, providing posterior, lateral, and medial views. Crucial points of interest include the alignment of the calcaneus with the talus and the navicular bone with the talus head, essential for proper foot correction. The dynamics of these markers were then extrapolated. Despite the constraints of a limited dataset, commendable repeatability in the results was attained, ensuring a certain reliability of measurements and robustness in the analysis. However, the intrinsic complexity of the Ponseti method, combined with potential variations in its execution, highlights the necessity to conduct a more extensive range of tests, considering various cases involving different surgeons. This approach will facilitate the development of a more comprehensive dataset, allowing for a more precise generalization of the Ponseti method across a wider spectrum of clinical scenarios.| File | Dimensione | Formato | |
|---|---|---|---|
|
2024_04_Miele_ExecutiveSummary_02.pdf
non accessibile
Descrizione: Testo Executive Summary
Dimensione
1.39 MB
Formato
Adobe PDF
|
1.39 MB | Adobe PDF | Visualizza/Apri |
|
2024_04_Miele_01.pdf
non accessibile
Descrizione: Testo Tesi
Dimensione
13.31 MB
Formato
Adobe PDF
|
13.31 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/10589/219855