Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Acceleration Signals In Determining Gait-Related Difficulties And The Motor Skill Of Walking In Older Adults

Dasgupta, Pritika (2021) Acceleration Signals In Determining Gait-Related Difficulties And The Motor Skill Of Walking In Older Adults. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 16 June 2022.

Download (14MB) | Request a Copy

Abstract

In adults 65 years or older, falls or other neuromotor dysfunctions are often framed as walking-related declines in motor skill; the frequent occurrence of such decline in walking-related motor skill motivates the need for an improved understanding of motor skill in walking. Simple gait measurements, such as speed, do not provide adequate information about the quality of the translation of the body motion during walking. Furthermore, there is a great need in the clinical literature and clinical practice for more accurate measures of the loss of the motor skill of walking, so that clinical practice can provide better therapeutic interventions to improve the motor skill of walking. This dissertation suggests a consensus on what the motor skill of walking is and dissects it into seven interrelated characteristics and traits. Subsequently, we purport that these characteristics of the motor skill of walking cannot be represented by simple gait measurements or raw sensor measurements alone. Gait measures from accelerometers placed on the lower trunk, or trunk-acceleration gait measures, can enrich measurements of walking and motor performance. To support our claim, we will map these acceleration gait measures (AGMs) to the various aspects of the motor skill of walking. Additionally, influential AGMs will be elected through feature selection methods. Various machine learning algorithms ranging from logistic regression, non-linear regression, evolutionary algorithms, and ensemble methods will be used to make predictions on age-related gait-related difficulty outcomes (such as fall risk). Overall, we hope to find that the combination of high-fidelity artificial intelligence algorithms and acceleration gait measures derived from low-cost sensors can fulfill the severe and crucial need for the clinical measurement of the motor skill of walking in older adults.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Dasgupta, Pritikapritika.dasgupta@gmail.comprd170000-0002-6199-2352
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairSejdić, Ervin
Committee MemberLu, Xinghua
Committee MemberVanSwearingen, Jessie
Committee MemberRedfern, Mark
Committee MemberTriantafillou, Sofia
Date: 16 June 2021
Date Type: Publication
Defense Date: 23 April 2021
Approval Date: 16 June 2021
Submission Date: 12 May 2021
Access Restriction: 1 year -- Restrict access to University of Pittsburgh for a period of 1 year.
Number of Pages: 165
Institution: University of Pittsburgh
Schools and Programs: School of Medicine > Biomedical Informatics
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: motor control, motor skill, movement control, acceleration, wearables, gait, clinical informatics, machine learning, feature specifi�cation
Date Deposited: 16 Jun 2021 16:51
Last Modified: 16 Jun 2021 16:51
URI: http://d-scholarship.pitt.edu/id/eprint/41076

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item