Fragility fractures due to falls in the elderly population reduce the quality of life and burden the health care system. A better understanding of the factors that discriminate injurious from non-injurious falls and the conditions that increase the risk of hip fracture is critical to improving screening techniques and developing effective treatments. Although muscle activation likely plays a role in fracture risk, very few studies investigate this topic. Existing work on muscle activation in finite element (FE) lateral fall models has often required considerable effort to choose muscle insertion points in the model-building process. In addition, the placement of insertion points is often subjective and, therefore, not reproducible. The purpose of this study is twofold: First: implement an automated method to derive both the muscle insertion points and the muscle force exerted of the individual specimens; Second: study the role of muscle contraction during lateral fall in fracture risk. The first part of the project was characterized by the use of the image registration technique through the Advanced Normalization Tools (ANTs) software. This allowed the computation of the coordinates of the muscle insertion points of each specimen. In addition, based on these results and the sample’s anthropometric data, an optimization problem to compute the force exerted by each muscle in absence of movement was solved. The second part involved the study of the role of muscle contraction in fracture risk. After implementing the muscles in the FE model, simulations were carried out, first according to a deterministic method, studying only four levels of muscle contraction (1,3,6 and 9 times body weight), and then according to a statistical method. The first analysis helped to understand the variation in contact forces between the acetabulum and femoral head; an increase in the level of contraction results in an increase in the force. In some cases the imposed muscle force was so high that the bone fracture occurred before the impact. These results helped the decision of the experiment design for the statistical method. Here, 220 simulations, designed according to the Latin Hypercube sampling technique, were evaluated for muscle activation values ranging between 1 and 5 times body weight and ground impact velocity values between 1.1 m/s and 3.5 m/s. This allowed us to compare the role of impact velocity, muscle activation level, and anthropometric characteristics of the subject in forces and deformations acting on the hip joint during a sideways fall. The results showed that all these parameters play a key role, and in particular, that muscle activation levels below a certain threshold, which varies depending on the specimen, have a protective effect on the fracture, as they result in better strain distribution. Furthermore, while an increase in ground force is observed in samples with higher body weight, the number of elements that yield, according to the fracture criterion, decreases, probably due to better energy absorption by thicker soft tissue. Statistical analysis applied to the finite element model showed that in 94.1% of the simulations, muscle activation plays a protective role. Except for rare cases in which fracture occurred before impact, the fracture event appears to be otherwise more related to impact velocity than to muscle contraction. Even in failed simulations, further studies are necessary to ensure that the fracture is due to contraction or if non-physiological values were considered. In the previous study in which the sideways fall simulations were performed without muscle inclusion and an impact velocity of 3.1 m/s, eight fractures out of eleven simulations were verified (72.72%). In this study, by contrast, 220 simulations were carried out, with varying values of impact velocity and muscle action. Thirty-six fractures occurred (16.36%), all in simulations with a high impact velocity. This study is part of a series aimed at increasing the scientific community’s knowledge of the influence of muscle activation concerning hip fracture risk. The main objective of the study was to provide a method that would allow easy and rapid implementation of muscles into the finite element model to simplify the inclusion of muscles in subsequent studies, even on a large population.
Le fratture da fragilità dovute a cadute nella popolazione anziana riducono la qualità della vita e gravano sul sistema sanitario. Una migliore comprensione dei fattori che dividono le cadute lesive da quelle non lesive e delle condizioni che aumentano il rischio di frattura dell’anca è fondamentale per migliorare le tecniche di screening e sviluppare trattamenti efficaci. Sebbene l’attivazione muscolare abbia probabilmente un ruolo nel rischio di frattura, ci sono pochissimi studi che indagano questo argomento. I lavori esistenti sull’attivazione muscolare nei modelli di caduta laterale a elementi finiti (FE) hanno spesso richiesto un notevole sforzo per scegliere i punti di inserzione muscolare nel processo di costruzione del modello. Inoltre, il posizionamento dei punti di inserzione è spesso soggettivo e, quindi, non riproducibile. Lo scopo di questo studio è duplice: In primo luogo, è necessario implementare un metodo automatizzato per ricavare i punti di inserzione muscolare dei singoli campioni e la forza esercitata. In secondo luogo, si vogliono implementare i muscoli con le caratteristiche specifiche del soggetto precedentemente derivate nel modello biofidelic agli elementi finiti di caduta laterale, al fine di studiare il loro ruolo nel rischio di frattura. La prima parte del progetto è stata caratterizzata dalla registrazione di immagini utilizzando il software Advanced Normalization Tools (ANTs). Ciò ha permesso di ricavare le coordinate dei punti di inserzione muscolare di ciascun campione. Inoltre, sulla base di questi risultati e dei dati antropometrici del campione, è stato risolto il problema dell’ottimizzazione della forza esercitata da ciascun muscolo in assenza di movimento. La seconda parte ha riguardato lo studio del ruolo della contrazione muscolare nel rischio di frattura. Dopo aver implementato i muscoli nel modello FE, sono state effettuate simulazioni, prima secondo un metodo deterministico, studiando solo 4 livelli di contrazione muscolare (1,3,6 e 9 volte il peso corporeo), e poi secondo un metodo statistico. La prima analisi ha permesso di comprendere la variazione della forza di contatto tra l’acetabolo e la testa del femore. Un aumento del livello di contrazione comporta un incremento di tale forza, talvolta così elevato da causare la frattura dell’osso prima dell’impatto. Questi risultati hanno aiutato a decidere il disegno dell’esperimento per il metodo statistico. Nello studio, 220 simulazioni, progettate secondo la tecnica di campionamento Latin Hypercube, sono state valutate per valori di attivazione muscolare compresi tra 1 e 5 volte il peso corporeo e valori di velocità di impatto al suolo compresi tra 1,1 m/s e 3,5 m/s. Questo ci ha permesso di confrontare il ruolo della velocità di impatto, del livello di attivazione muscolare e delle caratteristiche antropometriche del soggetto nella determinazione delle forze e delle deformazioni che agiscono sull’articolazione dell’anca durante una caduta laterale. I risultati hanno dimostrato che tutti questi parametri giocano un ruolo fondamentale, e in particolare che i livelli di attivazione muscolare al di sotto di una certa soglia, che varia a seconda del campione, hanno un effetto protettivo sulla frattura, in quanto determinano una migliore distribuzione delle deformazioni. Inoltre, mentre si osserva un aumento della forza al suolo nei campioni con peso corporeo maggiore, il numero di elementi che, secondo il criterio di frattura utilizzato, cedono diminuisce, probabilmente a causa di un migliore assorbimento dell’energia da parte di tessuti molli più spessi. L’analisi statistica applicata al modello a elementi finiti ha mostrato che nel 94, 1% delle simulazioni l’attivazione muscolare svolge un ruolo protettivo. Ad eccezione di rari casi in cui la frattura si è verificata prima dell’impatto, sembra essere altrimenti più legata alla velocità di impatto che alla contrazione muscolare. Inoltre, anche nelle simulazioni con fallimento osseo precoce, sono necessarie ulteriori valutazioni per valutare se i livelli di contrazione muscolare considerati rientrano in un range fisiologico. Questo studio fa parte di una serie volta ad aumentare le conoscenze della comunità scientifica sull’influenza dell’attivazione muscolare rispetto al rischio di frattura dell’anca. L’obiettivo principale dello studio era quello di fornire un metodo che consentisse una facile e rapida implementazione dei muscoli nel modello a elementi finiti, in modo da poter condurre studi su una popolazione più ampia.
Investigation of muscle co-contraction in hip fracture risk using a finite element sideways fall model
Maioli, Vera
2021/2022
Abstract
Fragility fractures due to falls in the elderly population reduce the quality of life and burden the health care system. A better understanding of the factors that discriminate injurious from non-injurious falls and the conditions that increase the risk of hip fracture is critical to improving screening techniques and developing effective treatments. Although muscle activation likely plays a role in fracture risk, very few studies investigate this topic. Existing work on muscle activation in finite element (FE) lateral fall models has often required considerable effort to choose muscle insertion points in the model-building process. In addition, the placement of insertion points is often subjective and, therefore, not reproducible. The purpose of this study is twofold: First: implement an automated method to derive both the muscle insertion points and the muscle force exerted of the individual specimens; Second: study the role of muscle contraction during lateral fall in fracture risk. The first part of the project was characterized by the use of the image registration technique through the Advanced Normalization Tools (ANTs) software. This allowed the computation of the coordinates of the muscle insertion points of each specimen. In addition, based on these results and the sample’s anthropometric data, an optimization problem to compute the force exerted by each muscle in absence of movement was solved. The second part involved the study of the role of muscle contraction in fracture risk. After implementing the muscles in the FE model, simulations were carried out, first according to a deterministic method, studying only four levels of muscle contraction (1,3,6 and 9 times body weight), and then according to a statistical method. The first analysis helped to understand the variation in contact forces between the acetabulum and femoral head; an increase in the level of contraction results in an increase in the force. In some cases the imposed muscle force was so high that the bone fracture occurred before the impact. These results helped the decision of the experiment design for the statistical method. Here, 220 simulations, designed according to the Latin Hypercube sampling technique, were evaluated for muscle activation values ranging between 1 and 5 times body weight and ground impact velocity values between 1.1 m/s and 3.5 m/s. This allowed us to compare the role of impact velocity, muscle activation level, and anthropometric characteristics of the subject in forces and deformations acting on the hip joint during a sideways fall. The results showed that all these parameters play a key role, and in particular, that muscle activation levels below a certain threshold, which varies depending on the specimen, have a protective effect on the fracture, as they result in better strain distribution. Furthermore, while an increase in ground force is observed in samples with higher body weight, the number of elements that yield, according to the fracture criterion, decreases, probably due to better energy absorption by thicker soft tissue. Statistical analysis applied to the finite element model showed that in 94.1% of the simulations, muscle activation plays a protective role. Except for rare cases in which fracture occurred before impact, the fracture event appears to be otherwise more related to impact velocity than to muscle contraction. Even in failed simulations, further studies are necessary to ensure that the fracture is due to contraction or if non-physiological values were considered. In the previous study in which the sideways fall simulations were performed without muscle inclusion and an impact velocity of 3.1 m/s, eight fractures out of eleven simulations were verified (72.72%). In this study, by contrast, 220 simulations were carried out, with varying values of impact velocity and muscle action. Thirty-six fractures occurred (16.36%), all in simulations with a high impact velocity. This study is part of a series aimed at increasing the scientific community’s knowledge of the influence of muscle activation concerning hip fracture risk. The main objective of the study was to provide a method that would allow easy and rapid implementation of muscles into the finite element model to simplify the inclusion of muscles in subsequent studies, even on a large population.File | Dimensione | Formato | |
---|---|---|---|
Master_Thesis_Maioli.pdf
accessibile in internet per tutti
Dimensione
31.33 MB
Formato
Adobe PDF
|
31.33 MB | Adobe PDF | Visualizza/Apri |
Executive_Summary_Maioli.pdf
accessibile in internet per tutti
Dimensione
5.4 MB
Formato
Adobe PDF
|
5.4 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/10589/189587