東華大學圖書館 |

Modeling and Predicting Response to Ankle Exoskeletons.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Modeling and Predicting Response to Ankle Exoskeletons./
Author:	Rosenberg, Michael C.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
Description:	175 p.
Notes:	Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
Contained By:	Dissertations Abstracts International83-02B.
Subject:	Biomechanics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28549361
ISBN:	9798535508344

Modeling and Predicting Response to Ankle Exoskeletons.
Rosenberg, Michael C.

Modeling and Predicting Response to Ankle Exoskeletons. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 175 p.

Source: Dissertations Abstracts International, Volume: 83-02, Section: B.

Thesis (Ph.D.)--University of Washington, 2021.

This item must not be sold to any third party vendors.

Ankle exoskeletons are designed and personalized to enhance mobility in unimpaired adults and improve gait in individuals with motor impairments. Ankle exoskeletons are challenging to prescribe and optimize for an individual, resulting in inconsistent intervention outcomes. Quantifying and predicting changes in kinematics and muscle activity in response to varying exoskeleton properties may improve intervention outcomes, enhance mobility, and inform device design for individuals with diverse motor impairments. However, model-based approaches for understanding responses to ankle exoskeletons often rely on physiologically-detailed frameworks that require extensive experimental datasets to model heterogeneous physiology and motor control in individuals with motor impairments. To better understand responses to ankle exoskeletons across individuals with diverse physiology, the goal of this dissertation is to examine physiologically-detailed and non-physiological approaches to modeling and understanding responses to ankle exoskeletons in individuals with motor impairments and unimpaired adults. Cerebral palsy (CP) is one of the most common motor impairments among children and one of the largest groups who use ankle exoskeletons. Optimized powered and passive ankle exoskeletons have been shown to reduce the energetic demands of walking in children with CP. However, CP represents a heterogeneous population, with widely varying gait patterns. Understanding how heterogeneous kinematics and kinetics in CP alter exoskeletons impact on the energetic demands of walking could inform device design. Using subject-specific musculoskeletal simulations of walking with ankle exoskeletons, we found that idealized powered ankle exoskeletons reduced muscle demand in children with CP more than passive exoskeletons and that reductions in ankle plantarflexor demand drove overall muscle demand. However, walking speed and knee flexion angle impacted reductions in muscle demand. Powered ankle exoskeletons may, therefore, benefit children with CP, but may not provide benefits over optimized passive exoskeletons for all individuals. While musculoskeletal simulations provide a powerful platform to evaluate 'what-if' questions and probe complex systems, the underlying models often have normative assumptions about physiology and motor control that may limit their ability to accurately predict subject-specific responses to exoskeletons. Constructing dynamical models of walking with exoskeletons from data alone may enable exoskeleton responses to be predicted without detailed knowledge of an individual's physiology. To test this theory, we built a passive ankle exoskeleton and collected extended treadmill walking data across four levels of dorsiflexion stiffness for 12 unimpaired adults. We developed three data-driven phase-varying models of each individual's response to ankle exoskeleton torque and evaluated their predictive ability in unimpaired adults walking in bilateral ankle exoskeletons. We found that linear and nonlinear phase-varying models could accurately predict kinematic responses to torque but could not predict stride-to-stride variations in myoelectric responses. These models show promising potential to model responses to exoskeletons in individuals with motor impairments, though improving myoelectric predictions represents an exciting area of future research. While exoskeleton impacts on gait mechanics and energetics have been investigated, if and how an individual modulates their center-of-mass (COM) dynamics changes with ankle exoskeletons remains unclear. Quantifying changes in the whole-limb mechanisms describing COM dynamics with exoskeletons may identify characteristic sub-classes of responses to exoskeletons. We developed and identified template signatures - low-dimensional, physics-based representations of COM dynamics - during walking with and without passive ankle exoskeletons in 12 unimpaired adults and one individual with post-stroke hemiparesis. We found that the template signatures were consistent across unimpaired adults and were robust to changes in exoskeleton dorsiflexion stiffness. Conversely, the template signatures post-stroke reflected the individual's increased paretic-limb stiffness and changed in response to exoskeletons. This work suggests that unimpaired COM dynamics do not change with passive ankle exoskeletons, but that COM dynamics in individuals post-stroke may adapt to ankle exoskeletons.This dissertation contributed to our knowledge of how ankle exoskeleton properties impact muscle demand and COM dynamics during walking, and the potential of data-driven modeling frameworks to quantify and predict responses to ankle exoskeletons. This knowledge may inform exoskeleton design and prescription, potentially improving exoskeleton efficacy for individuals with motor impairments. The research in this dissertation added to an existing open-source dataset for musculoskeletal simulations of walking in children with CP and created a novel, open-source dataset containing long time-series of walking with passive ankle exoskeletons in unimpaired adults. This research lays the foundation for future work aimed at identifying mechanisms driving heterogeneous responses to exoskeletons or other assistive devices in individuals with motor impairments.

ISBN: 9798535508344Subjects--Topical Terms:

548685
Biomechanics.
Subjects--Index Terms:

Assistive device

Modeling and Predicting Response to Ankle Exoskeletons.
LDR:06523nmm a2200397 4500 001 2351901
005 20221111113709.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798535508344
035 $a (MiAaPQ)AAI28549361
035 $a AAI28549361
040 $a MiAaPQ $c MiAaPQ
100 1 $a Rosenberg, Michael C. $3 3691490
245 1 0 $a Modeling and Predicting Response to Ankle Exoskeletons.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 175 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
500 $a Advisor: Steele, Katherine M.
502 $a Thesis (Ph.D.)--University of Washington, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a Ankle exoskeletons are designed and personalized to enhance mobility in unimpaired adults and improve gait in individuals with motor impairments. Ankle exoskeletons are challenging to prescribe and optimize for an individual, resulting in inconsistent intervention outcomes. Quantifying and predicting changes in kinematics and muscle activity in response to varying exoskeleton properties may improve intervention outcomes, enhance mobility, and inform device design for individuals with diverse motor impairments. However, model-based approaches for understanding responses to ankle exoskeletons often rely on physiologically-detailed frameworks that require extensive experimental datasets to model heterogeneous physiology and motor control in individuals with motor impairments. To better understand responses to ankle exoskeletons across individuals with diverse physiology, the goal of this dissertation is to examine physiologically-detailed and non-physiological approaches to modeling and understanding responses to ankle exoskeletons in individuals with motor impairments and unimpaired adults. Cerebral palsy (CP) is one of the most common motor impairments among children and one of the largest groups who use ankle exoskeletons. Optimized powered and passive ankle exoskeletons have been shown to reduce the energetic demands of walking in children with CP. However, CP represents a heterogeneous population, with widely varying gait patterns. Understanding how heterogeneous kinematics and kinetics in CP alter exoskeletons impact on the energetic demands of walking could inform device design. Using subject-specific musculoskeletal simulations of walking with ankle exoskeletons, we found that idealized powered ankle exoskeletons reduced muscle demand in children with CP more than passive exoskeletons and that reductions in ankle plantarflexor demand drove overall muscle demand. However, walking speed and knee flexion angle impacted reductions in muscle demand. Powered ankle exoskeletons may, therefore, benefit children with CP, but may not provide benefits over optimized passive exoskeletons for all individuals. While musculoskeletal simulations provide a powerful platform to evaluate 'what-if' questions and probe complex systems, the underlying models often have normative assumptions about physiology and motor control that may limit their ability to accurately predict subject-specific responses to exoskeletons. Constructing dynamical models of walking with exoskeletons from data alone may enable exoskeleton responses to be predicted without detailed knowledge of an individual's physiology. To test this theory, we built a passive ankle exoskeleton and collected extended treadmill walking data across four levels of dorsiflexion stiffness for 12 unimpaired adults. We developed three data-driven phase-varying models of each individual's response to ankle exoskeleton torque and evaluated their predictive ability in unimpaired adults walking in bilateral ankle exoskeletons. We found that linear and nonlinear phase-varying models could accurately predict kinematic responses to torque but could not predict stride-to-stride variations in myoelectric responses. These models show promising potential to model responses to exoskeletons in individuals with motor impairments, though improving myoelectric predictions represents an exciting area of future research. While exoskeleton impacts on gait mechanics and energetics have been investigated, if and how an individual modulates their center-of-mass (COM) dynamics changes with ankle exoskeletons remains unclear. Quantifying changes in the whole-limb mechanisms describing COM dynamics with exoskeletons may identify characteristic sub-classes of responses to exoskeletons. We developed and identified template signatures - low-dimensional, physics-based representations of COM dynamics - during walking with and without passive ankle exoskeletons in 12 unimpaired adults and one individual with post-stroke hemiparesis. We found that the template signatures were consistent across unimpaired adults and were robust to changes in exoskeleton dorsiflexion stiffness. Conversely, the template signatures post-stroke reflected the individual's increased paretic-limb stiffness and changed in response to exoskeletons. This work suggests that unimpaired COM dynamics do not change with passive ankle exoskeletons, but that COM dynamics in individuals post-stroke may adapt to ankle exoskeletons.This dissertation contributed to our knowledge of how ankle exoskeleton properties impact muscle demand and COM dynamics during walking, and the potential of data-driven modeling frameworks to quantify and predict responses to ankle exoskeletons. This knowledge may inform exoskeleton design and prescription, potentially improving exoskeleton efficacy for individuals with motor impairments. The research in this dissertation added to an existing open-source dataset for musculoskeletal simulations of walking in children with CP and created a novel, open-source dataset containing long time-series of walking with passive ankle exoskeletons in unimpaired adults. This research lays the foundation for future work aimed at identifying mechanisms driving heterogeneous responses to exoskeletons or other assistive devices in individuals with motor impairments.
590 $a School code: 0250.
650 4 $a Biomechanics. $3 548685
650 4 $a Biomedical engineering. $3 535387
650 4 $a Bioinformatics. $3 553671
650 4 $a Ankle. $3 3563106
650 4 $a Research. $3 531893
650 4 $a Physiology. $3 518431
650 4 $a Mechanical properties. $3 3549505
650 4 $a Kinematics. $3 571109
650 4 $a Datasets. $3 3541416
650 4 $a Legs. $3 3563163
650 4 $a Dissertations & theses. $3 3560115
650 4 $a Metabolism. $3 541349
650 4 $a Fitness equipment. $3 3563152
650 4 $a Dynamical systems. $3 3219613
650 4 $a System theory. $3 525574
650 4 $a Gait. $3 1362772
650 4 $a Simulation. $3 644748
650 4 $a Stroke. $3 1362725
650 4 $a Physics. $3 516296
650 4 $a Engineering schools. $3 3691491
650 4 $a Variables. $3 3548259
650 4 $a Cerebral palsy. $3 894315
650 4 $a Muscle function. $3 3562993
650 4 $a Walking. $3 548733
650 4 $a Posture. $3 667530
650 4 $a Adults. $3 2157228
650 4 $a Scholarships & fellowships. $3 3560311
653 $a Assistive device
653 $a Cerebral palsy
653 $a Data-driven modeling
653 $a Exoskeleton
653 $a Locomotion
690 $a 0648
690 $a 0541
690 $a 0715
690 $a 0605
690 $a 0719
710 2 $a University of Washington. $b Mechanical Engineering. $3 2092993
773 0 $t Dissertations Abstracts International $g 83-02B.
790 $a 0250
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28549361