Back to search

University of Strathclyde

Data science enabled rehabilitation

Abstract

dc:description.abstract

Stroke is a main cause of impairment/disability. More stroke survivors undergounsupervised home rehabilitation. Autonomous self-rehabilitation systems usingsensing and machine learning are not tailored to patients’ needs.Based on a systematic narrative literature review, home-based rehabilitation systemswere taxonomized and new design criteria were formulated for increased patientengagement enhancement and individualism. No system that addresses all the criteriawas found in literature. An in-house low-cost home-based rehabilitation AmbientIntelligence (AmI) system was deployed meeting the criteria, and an accuracyevaluation method proposed, in line with medically approved tests. The Timed Up andGo (TUG) and Five Time Sit To Stand (FTSTS) tests evaluate daily living activityperformance in the presence/development of comorbidities. The AmI-driven systemcomplies with Accountability, Responsibility, and Transparency (ART) requirementsfor wider acceptability. A method is presented for generating synthetic datasetscomplementing experimental observations mitigating bias present due to practicallimitations. Also, an incremental hybrid machine learning algorithm is proposed. Itcombines ensemble learning and hybrid stacking using extreme gradient boosteddecision trees and k-nearest neighbours to meet individualisation, and ARTrequirements while maintaining low computation footprint.The proposed approach was based on the criteria: nonintrusive, nonwearable,motivation and engagement enhancing, individualized, supporting daily activities,cost-effective, simple, and transferable. The motivation method, suitability for elderly,and intended use were examined as supplementary criteria. Indicators of enhancedmotivation and engagement, through questionnaire responses, demonstrate that >83%of participants support the proposed system’s motivation and engagementenhancement. The system is fit for purpose with statistically significant (?c>0.99,R2>0.94, ICC>0.96) and unbiased correlation to the gold standard. The model reachesup to 100% accuracy for FTSTS and TUG in predicting associated patient medicalcondition, and 100% or 83.13%, respectively, in predicting area of difficulty in thesegments of the test. Results show an improvement of 5% and 15% for FTSTS andTUG, over previous intrusive approaches.Keywords:Home-based rehabilitation systems, Stroke rehabilitation, Telerehabilitation, Patientparticipation, Motivation, Comparative effectiveness research, Automated timed upand go test, Automated five time sit to stand test, Self-evaluation, Evaluation ofsensor systems, Non-intrusive sensing, Sensing for health, Accountable ArtificialIntelligence, Responsible Artificial Intelligence, Transparent Artificial Intelligence,Hybrid ensemble learning, Patient-centric individualised rehabilitation

Degree

thesis:*
Name dc:type.qualificationname
phd
Level dc:type.qualificationlevel
doctoral-pg
Grantor dc:publisher.institution
University of Strathclyde
Year dc:date.issued
2020

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Vourganas, Ioannis
Advisor dc:contributor.advisor
  • Stanković, Vladimir
Author Identifier dc:creator.authoridentifier
  • 201772434

Rights

Identifiers

dc:identifier.*
Identifier
T15869

Chain of custody

source
Harvested from
University of Strathclyde
Base URL
stax.strath.ac.uk/catalog/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Vourganas, Ioannis. Data science enabled rehabilitation. doctoral-pg thesis, University of Strathclyde, 2020. https://stax.strath.ac.uk/concern/theses/b5644r54d