The present work, carried out at the laboratory of Movement Biomechanics and Motor Control (MBMC Lab) of the Politecnico di Milano, aims at developing a new method to calculate power and energy flow at the ankle joint. The standard method currently used in literature, named in this work the “Ankle Joint method” (AJ), approximates the foot as a rigid body. This approximation might strongly influence the power estimation. Indeed, the power transmitted to or absorbed by the shank is calculated considering only the relative rotation of the foot with respect to the shank, as the product between the joint moment and the angular velocity of the foot with respect to the ankle joint axis. The method developed in our work, called the “Deformable Foot method” (DF), evaluates the power transmitted to the shank as the sum of two different terms: rotational and translational. In this case, power calculation intrinsically considers the effects of foot deformations and rotations around other axis than the flexion/extension one. In particular, the rotational term is equal to the product of the ankle joint moment and the angular velocity of the shank. The translational power is due to the ground reaction force applied to the ankle joint and the linear velocity of the ankle joint centre. Once the power is calculated with the two methods, the energy flow is obtained by integrating power over time. Some approximations are underlined also by the new method (DF). In particular, the inertial forces and moments of the foot are neglected. In a deformable segment, the identification of the center of mass and moments of inertia requires a complex analysis of the different configurations assumed during the deformation. This evaluation would require a complicated markers placement protocol, able to describe the various deformations of the foot during the stance phase. However, considering that the foot accelerations are very low during the stance phase and the foot mass is relatively small in relation to the whole body mass, we decided to discard the inertial components, definitely negligible with respect to the ground reaction forces and moments. Our method provides therefore a reliable power and energy estimation during gait without increasing the complexity of application (low number of markers). We acquired kinematic and dynamic data of the left and right limbs of eight healthy subjects and two transtibial amputee subjects, by means of a motion analysis system (Smart-E, BTS, Italy) with 6 cameras (sampling frequency of 120 Hz), and a force platform (sampling frequency of 960 Hz, Kistler, 9286AA, Switzerland). In particular, the trials for the 8 healthy subjects were performed in barefoot condition, at three different walking speeds (slow, normal and high) and with sport shoes at normal speed. For the female subjects, we also added a walking test in high heels at normal speed. The walking trials for the two amputees were performed with and without shoes at normal speed. We used a marker protocol with six passive markers, placed on the two metatarsi (first and fifth), the two malleoli (medial and lateral), the medial and lateral condyle. We positioned the markers on the prosthetic limb on points correspondent to those identified for the sound limb. The first analysis considered the healthy subjects group. The comparison of the power curves obtained with the two methods revealed significant differences: the DF method showed an absorption negative peak during the first part of the stance phase, which is not visible with the AJ method; the maximum power value calculated with the DF method was lower in almost all trials with respect to the one calculated with the AJ method. These differences were present in all conditions (at different velocities, barefoot walking, walking with shoes, in high heels). The results suggest that the AJ method overestimates the power generated, and consequently the plantar-flexor muscles activity. With regards to the gait velocity, the final values of energy obtained with the AJ and DF methods were analysed. With the AJ method, the final energy value was always positive; with the DF method, instead, we obtained negative values at low and normal walking speed, and positive at high speed. These results underline the inability of the AJ method to provide a reliable evaluation of the energy flows between foot and shank, since it does not consider foot’s structure deformations, which involve energy dissipation. The value of final energy and maximum peak of power showed a statistically significant difference for all three walking speeds (Wilcoxon test, p-value <0.05), confirming that the velocity is a determinant factor for the analysis of power and energy curves. Comparing the barefoot walking and the walking with sport shoes, we showed that the produced power is lower in the latter condition, possibly due to the soft and viscoelastic materials of the shoes. Regarding the walking in high heels, female healthy subjects showed changes in power and energy curves probably related to the altered gait pattern and muscular activity. The second part of our study was focused on two transtibial amputee subjects. Concerning the prosthetic limbs, the Ankle Joint method was not applied because the artificial foot has no joints at the ankle, and the model of rigid foot is not valid. The final values of the transmitted energy from the prosthetic foot to the pylon were always found negative because a part of the absorbed energy is dissipated, and during the phase of energy production, the amputee subjects cannot rely on the contribution of the plantar-flexor muscles, which guarantees in healthy subjects a higher final energy value. After analysing the performance of the prosthetic foot, we also evaluated the power and energy profiles of the sound limb, to investigate if compensatory mechanisms/alterations due to the use of the prosthesis were present in the contralateral limb. We first selected for each amputee subject a velocity-matched healthy control group because of the great influence of velocity in power calculation. The final energy value and the maximum peak of power were not statistically different across genders in the healthy control group (Wilxon test, p-value>0.05). We therefore included both genders in the control groups definition for the comparison with the amputee subjects. The results of the comparison showed similar behaviour between the sound limb of the amputee subjects and the healthy control groups, showing that the presence of the prosthesis was not affecting the walking pattern of the contralateral limb. The evaluation of power and energy exchanged at the foot has an important role in supporting and assisting choices needed to improve the quality of life of those who live with functional limitations. Therefore, a correct and accurate calculation of these quantities is required. These preliminary results are the basis for more ambitious studies aimed at the evaluation of power and energy of the foot in different pathological conditions to address clinicians towards most effective decisions.
Il presente lavoro di tesi, svolto presso il laboratorio MBMC del Politecnico di Milano, mira allo sviluppo di un metodo per calcolare potenza e scambi energetici all’articolazione della caviglia. Il metodo che viene generalmente utilizzato negli studi di analisi del movimento, denominato in questo lavoro “Ankle Joint method” (AJ), modellizza il piede come un corpo rigido, senza considerarne le deformazioni presenti a livello fisiologico durante il cammino. Tale approssimazione influenza in maniera importante il calcolo della potenza trasmessa o assorbita dalla tibia, che viene ricavata considerando solo il movimento relativo del piede rispetto alla tibia ed espressa come il prodotto tra il momento e la velocità angolare del piede attorno all'asse di rotazione della caviglia. Il metodo sviluppato nel nostro lavoro, chiamato “Deformable Foot method” (DF), valuta invece la potenza trasmessa alla tibia come la somma di due differenti termini: rotazionale e traslazionale. In questo caso gli effetti delle deformazioni del piede e delle rotazioni attorno ad altri assi oltre quello di flessione/estensione sono intrinsecamente considerati. In particolare, il termine della potenza associato alla rotazione è uguale al prodotto del momento dell'articolazione della caviglia per la velocità angolare della tibia intorno ai tre assi. La potenza traslazionale è dovuta invece alla forza di reazione al terreno applicata all'articolazione della caviglia e alla velocità lineare del centro dell'articolazione della caviglia. Una volta calcolate le potenze con i due metodi, il flusso di energia è stato ottenuto dall'integrazione della potenza nel tempo. Alcune approssimazioni nella modellizzazione del piede sono implicite anche nel nuovo metodo (DF). In particolare, le forze e i momenti di inerzia del piede vengono trascurati. Infatti, in un segmento deformabile l'identificazione del centro di massa e dei momenti di inerzia richiederebbe l’analisi delle diverse configurazioni assunte dal corpo durante la deformazione. Per effettuare questo tipo di valutazione sarebbe dunque necessaria una markerizzazione molto complessa che consenta la descrizione delle deformazioni del piede sotto carico. Tuttavia, visto e considerato che le accelerazioni del piede sono molto basse durante tale fase e la massa del piede è relativamente piccola in relazione all'intera massa corporea, abbiamo ritenuto le componenti inerziali trascurabili rispetto alle forze e ai momenti di reazione al terreno. Il metodo da noi utilizzato permette di calcolare l’energia associata alla deformazione del piede senza necessità di porre marcatori su di esso. Risulta quindi di facile applicazione (numero limitato di marcatori) anche in condizioni in cui il piede sia coperto dalla calzatura. Abbiamo raccolto dati cinematici e dinamici del cammino di 8 soggetti sani e due soggetti amputati transtibiali, sia per l'arto destro sia per quello sinistro, attraverso un sistema di analisi del movimento (Smart-E, BTS, Italia) con 6 telecamere (frequenza di acquisizione di120 Hz) e una piattaforma dinamometrica (Kistler, 9286AA, Svizzera, frequenza di acquisizione di 960 Hz,). In particolare, le prove per gli 8 soggetti sani sono state condotte a piedi nudi a tre diverse velocità (lenta, normale e alta) e con scarpe sportive solo a velocità normale. Per le donne è stato anche aggiunto un test di camminata con tacchi alti a velocità normale. Le prove per i due amputati sono invece state effettuate con e senza scarpe a velocità normale. Abbiamo utilizzato un protocollo di markerizzazione a sei marcatori passivi, posizionati sui due metatarsi (primo e quinto), i due malleoli (interno ed esterno), il condilo laterale e mediale. Per i soggetti amputati, i marker sull'arto protesico sono stati posizionati in punti corrispondenti a quelli identificati per l’arto sano. La prima analisi è stata condotta considerando i soggetti sani. Il confronto delle curve di potenza tra i due metodi ha rivelato differenze significative: il metodo DF mostra un picco negativo di assorbimento nella prima parte della stance phase, non visibile con il metodo AJ; il massimo valore di potenza nel metodo DF è inferiore nella maggioranza delle prove rispetto a quello di AJ. Queste differenze vengono mantenute per tutte le condizioni (velocità diverse, cammino a piedi nudi, con scarpe sportive, con tacchi alti). I risultati mostrano che il metodo AJ sovrastimi la potenza generata, sopravvalutando l'attività dei muscoli flessori plantari. Per quanto riguarda la velocità, vengono analizzati i valori finali dell'energia ottenuta con i metodi AJ e DF. Con il metodo AJ, vengono ottenuti valori sempre positivi, mentre con il metodo DF l’energia finale risulta sempre negativa per velocità di deambulazione bassa e normale, positiva per il cammino ad alta velocità. Questi risultati evidenziano come il metodo AJ non sia in grado di stimare correttamente i flussi reali di energia fra piede e tibia dal momento che non considera le deformazioni della struttura del piede, che comportano una dissipazione di energia. Il valore dell'energia finale e il picco massimo di potenza hanno mostrato una differenza statisticamente significativa fra le tre velocità di camminata (test di Wilcoxon, valore p <0,05), confermando come la velocità sia un fattore determinante per l'analisi delle curve di potenza e di energia. I risultati del confronto tra cammino a piedi nudi e scarpe sportive hanno dimostrato che la potenza prodotta durante la deambulazione con scarpe sportive diminuisce, probabilmente a causa delle loro caratteristiche morbide e viscoelastiche. Nei soggetti sani di sesso femminile il cammino con i tacchi ha mostrato cambiamenti nelle curve di potenza e di energia probabilmente correlati ad una alterazione nel normale pattern di cammino e nell’attività muscolare. La seconda parte del nostro studio è stata incentrata sui due soggetti amputati transtibiali. Per quanto riguarda gli arti protesici, il metodo AJ non è stato applicato perché il piede artificiale non possedendo articolazioni alla caviglia annulla la validità del modello del piede rigido. I valori finali dell'energia trasmessa dal piede protesico al pilone sono sempre risultati negativi perché una parte dell'energia assorbita viene dissipata e durante la fase di produzione di energia, i soggetti amputati non possono contare sul contributo dei muscoli flessori plantari, che garantisce nei soggetti sani un valore energetico finale più elevato. Dopo aver analizzato le prestazioni del piede protesico, abbiamo deciso di valutare anche l’arto sano, per verificare se fossero presenti meccanismi compensatori e/o alterazioni nell’andamento di potenza ed energia dati dalla presenza della protesi. Abbiamo in primo luogo selezionato per ognuno dei due soggetti un gruppo di controllo appropriato in termini di velocità di cammino, vista la sua grande influenza nel determinare la potenza. Il valore dell'energia finale e il picco massimo di potenza non sono risultati differenti fra femmine e maschi sani (test di Wilcoxon, valore p> 0,05), quindi entrambi i generi sono stati inclusi nella composizione dei gruppi di controllo. I risultati del confronto hanno mostrato un comportamento simile tra l'arto sano dell'amputato e il gruppo di controllo, mostrando che la presenza della protesi non andava ad alterare il pattern di cammino dell’arto controlaterale. La valutazione dei dati relativi a potenza ed energia svolge un ruolo importante nel supportare e assistere le scelte necessarie per migliorare la qualità della vita di coloro che vivono con limitazioni funzionali, pertanto è richiesto un loro calcolo corretto e accurato. Questi risultati preliminari sono la base per uno studio più ambizioso volto ad una migliore valutazione della potenza e dell'energia a livello del piede in diverse condizioni patologiche per indirizzare i clinici verso le decisioni più efficaci.
Methods for evaluating power and energy at the foot during gait for healthy and amputee subjects
FARINELLI, VERONICA;HOSSEINZADEH, LADAN
2017/2018
Abstract
The present work, carried out at the laboratory of Movement Biomechanics and Motor Control (MBMC Lab) of the Politecnico di Milano, aims at developing a new method to calculate power and energy flow at the ankle joint. The standard method currently used in literature, named in this work the “Ankle Joint method” (AJ), approximates the foot as a rigid body. This approximation might strongly influence the power estimation. Indeed, the power transmitted to or absorbed by the shank is calculated considering only the relative rotation of the foot with respect to the shank, as the product between the joint moment and the angular velocity of the foot with respect to the ankle joint axis. The method developed in our work, called the “Deformable Foot method” (DF), evaluates the power transmitted to the shank as the sum of two different terms: rotational and translational. In this case, power calculation intrinsically considers the effects of foot deformations and rotations around other axis than the flexion/extension one. In particular, the rotational term is equal to the product of the ankle joint moment and the angular velocity of the shank. The translational power is due to the ground reaction force applied to the ankle joint and the linear velocity of the ankle joint centre. Once the power is calculated with the two methods, the energy flow is obtained by integrating power over time. Some approximations are underlined also by the new method (DF). In particular, the inertial forces and moments of the foot are neglected. In a deformable segment, the identification of the center of mass and moments of inertia requires a complex analysis of the different configurations assumed during the deformation. This evaluation would require a complicated markers placement protocol, able to describe the various deformations of the foot during the stance phase. However, considering that the foot accelerations are very low during the stance phase and the foot mass is relatively small in relation to the whole body mass, we decided to discard the inertial components, definitely negligible with respect to the ground reaction forces and moments. Our method provides therefore a reliable power and energy estimation during gait without increasing the complexity of application (low number of markers). We acquired kinematic and dynamic data of the left and right limbs of eight healthy subjects and two transtibial amputee subjects, by means of a motion analysis system (Smart-E, BTS, Italy) with 6 cameras (sampling frequency of 120 Hz), and a force platform (sampling frequency of 960 Hz, Kistler, 9286AA, Switzerland). In particular, the trials for the 8 healthy subjects were performed in barefoot condition, at three different walking speeds (slow, normal and high) and with sport shoes at normal speed. For the female subjects, we also added a walking test in high heels at normal speed. The walking trials for the two amputees were performed with and without shoes at normal speed. We used a marker protocol with six passive markers, placed on the two metatarsi (first and fifth), the two malleoli (medial and lateral), the medial and lateral condyle. We positioned the markers on the prosthetic limb on points correspondent to those identified for the sound limb. The first analysis considered the healthy subjects group. The comparison of the power curves obtained with the two methods revealed significant differences: the DF method showed an absorption negative peak during the first part of the stance phase, which is not visible with the AJ method; the maximum power value calculated with the DF method was lower in almost all trials with respect to the one calculated with the AJ method. These differences were present in all conditions (at different velocities, barefoot walking, walking with shoes, in high heels). The results suggest that the AJ method overestimates the power generated, and consequently the plantar-flexor muscles activity. With regards to the gait velocity, the final values of energy obtained with the AJ and DF methods were analysed. With the AJ method, the final energy value was always positive; with the DF method, instead, we obtained negative values at low and normal walking speed, and positive at high speed. These results underline the inability of the AJ method to provide a reliable evaluation of the energy flows between foot and shank, since it does not consider foot’s structure deformations, which involve energy dissipation. The value of final energy and maximum peak of power showed a statistically significant difference for all three walking speeds (Wilcoxon test, p-value <0.05), confirming that the velocity is a determinant factor for the analysis of power and energy curves. Comparing the barefoot walking and the walking with sport shoes, we showed that the produced power is lower in the latter condition, possibly due to the soft and viscoelastic materials of the shoes. Regarding the walking in high heels, female healthy subjects showed changes in power and energy curves probably related to the altered gait pattern and muscular activity. The second part of our study was focused on two transtibial amputee subjects. Concerning the prosthetic limbs, the Ankle Joint method was not applied because the artificial foot has no joints at the ankle, and the model of rigid foot is not valid. The final values of the transmitted energy from the prosthetic foot to the pylon were always found negative because a part of the absorbed energy is dissipated, and during the phase of energy production, the amputee subjects cannot rely on the contribution of the plantar-flexor muscles, which guarantees in healthy subjects a higher final energy value. After analysing the performance of the prosthetic foot, we also evaluated the power and energy profiles of the sound limb, to investigate if compensatory mechanisms/alterations due to the use of the prosthesis were present in the contralateral limb. We first selected for each amputee subject a velocity-matched healthy control group because of the great influence of velocity in power calculation. The final energy value and the maximum peak of power were not statistically different across genders in the healthy control group (Wilxon test, p-value>0.05). We therefore included both genders in the control groups definition for the comparison with the amputee subjects. The results of the comparison showed similar behaviour between the sound limb of the amputee subjects and the healthy control groups, showing that the presence of the prosthesis was not affecting the walking pattern of the contralateral limb. The evaluation of power and energy exchanged at the foot has an important role in supporting and assisting choices needed to improve the quality of life of those who live with functional limitations. Therefore, a correct and accurate calculation of these quantities is required. These preliminary results are the basis for more ambitious studies aimed at the evaluation of power and energy of the foot in different pathological conditions to address clinicians towards most effective decisions.File | Dimensione | Formato | |
---|---|---|---|
Thesis Farinelli Hosseinzadeh.pdf
non accessibile
Descrizione: methods for the evaluation of power and energy at the foot during gait for healthy and amputee subjects
Dimensione
5.69 MB
Formato
Adobe PDF
|
5.69 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/10589/140217