<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T01:51:41Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/33097" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/33097</identifier><datestamp>2026-09-24T06:37:35Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Fang, Qiyin</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Jeff Gajdacs</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department">Engineering Physics</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2026-05-05T19:43:40Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2026</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://hdl.handle.net/11375/33097</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/15097</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract">The development of accessible, quantitative tools for remote physiotherapy monitoring has become an urgent priority in light of global disparities in rehabilitation access and the growing demand for decentralized healthcare delivery. This thesis presents the design, implementation, and validation of the Smart Exercise System, a privacy-preserving vision-based platform for two- and three-dimensional kinematic analysis of therapeutic exercises. The system repurposes commercially available smart camera, originally designed for fall detection, into a stereoscopic motion capture configuration. By leveraging on-device pose estimation that transmits only anonymized skeletal coordinates, the system achieves privacy-by-design, enabling deployment in private homes without the surveillance concerns accompanying continuous video monitoring.
A custom software pipeline was developed to retrieve skeletal data from the cloud, reconstruct three-dimensional joint positions from orthogonal camera views, calculate joint angles using vector mathematics, and derive secondary metrics including angular velocity and acceleration. The pipeline implements data quality management through missing frame logging, short-gap interpolation, and Savitzky-Golay filtering to reduce noise while preserving biomechanically relevant signal features.
Experimental validation characterized system performance across distances, environmental conditions, and reference standards. The 12-foot distance yielded optimal tracking with average positional standard deviation of (10.6mm) and dropout rate of 0.36 missing nodes per second. The system operates at a nominal framerate of approximately 5 frames per second, limiting its applicability to slow, controlled therapeutic exercises rather than rapid movements. Stress-testing demonstrated robustness to variable clothing and dim lighting, with no statistically significant degradation in tracking accuracy. A modified black background calibration was validated as having no effect on measurement accuracy while enhancing privacy and visual contrast. Direct comparison against manual goniometry confirmed strong agreement for static joint angle measurements, and alignment with inertial measurement unit data confirmed accurate temporal capture of movement phases.
The system's limitations include low framerate, dependency on proprietary cloud infrastructure, and incompatibility with enterprise Wi-Fi networks. Despite these constraints, the Smart Exercise System provides sufficiently accurate and reliable kinematic data for assessing gross motor patterns, joint range of motion, and movement symmetry in controlled therapeutic exercises. The characterization of its operational boundaries establishes the evidence base for informed deployment decisions. Future work includes clinical validation with patient populations, transition to open-source pose estimation frameworks, integration of hybrid sensing with inertial measurement units, and development of standardized movement databases to support equitable, data-driven rehabilitation.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="sponsorship">Natural Sciences and Engineering Research Council of Canada (NSERC)</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree">Master of Applied Science (MASc)</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso">en</dim:field>
   <dim:field mdschema="dc" element="rights" lang="en">Attribution 2.5 Canada</dim:field>
   <dim:field mdschema="dc" element="rights" qualifier="uri">http://creativecommons.org/licenses/by/2.5/ca/</dim:field>
   <dim:field mdschema="dc" element="title">A Vision-Based Measurement System for Kinematic Analysis in Remote Physiotherapy Applications</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">embargo</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>