Posters
Presenting Author Academic/Professional Position
MS2
Academic Level (Author 1)
Medical Student
Academic Level (Author 2)
Faculty
Discipline/Specialty (Author 2)
Neuroscience
Discipline Track
Biomedical ENGR/Technology/Computation
Abstract Type
Research/Clinical
Abstract
Background: Marker-less motion capture systems have improved the assessment of human movement by allowing for a non-invasive, affordable, quantitative data collection useful in clinical settings. However, their use in real-world community care settings is largely underexplored, presenting a unique opportunity to bridge the gap between controlled laboratory environments and the complexity of providing care in diverse, everyday contexts. This systematic review examines the differential use of MMC systems across real-world indoor environments (clinics, homes, living labs) and outdoor settings (public walkways, community spaces), with a focus on their application to neurological disorders such as stroke and neurodegenerative diseases.
Methods: A systematic review was conducted using databases including PubMed, Web of Science, MEDLINE (NIH) and IEEE for studies published since 2010. Search terms utilized included "marker-less motion capture," "indoor," "outdoor" "gait" "neurodegeneration," and related phrases. The initial search yielded 63 articles. After removing duplicates, 41 articles remained for screening. Following a detailed review of titles and abstracts against predefined inclusion and exclusion criteria, 7 articles were selected for the final full-text analysis. The papers were analyzed to identify key theme of indoor versus outdoor applications.
Results: Indoor MMC systems demonstrated high reliability and ease of integration in home and clinical settings. Gait velocity measurements yielded ICCs between 0.88 and 0.94, and step length errors were less than 2.5 cm. Upper limb motion capture achieved joint angle ICCs of ~0.87. Usability of these systems were high for both patients and clinicians with scores exceeding 80/100. These systems are also able to be integrated with telehealth platforms, offering consistent data capture under controlled lighting and spatial conditions. In contrast, outdoor MMC systems provided the advantage of ecological validity but showed decreased precision. Joint tracking mean absolute errors ranged from 4.2 to 6.7 cm, and lighting variability, occlusions, and background interference contributed to a 15–25% reduction in data fidelity. Temporal alignment issues resulted in errors up to 100 ms.
Conclusion: Indoor MMC systems currently offer improved measurement accuracy, reliability, and user integration for neurorehabilitation applications, making them more feasible for clinical and home-based deployment. Outdoor systems, while technically limited, provide unique insights into real-world movement patterns that are essential for functional outcome assessments. Assessing the improvements of computer vision, AI-driven occlusion, and sensor fusion can help bridge the gap between the systems. Future research should focus on longitudinal validation of these systems with comparison to clinical metrics to improve equity and inform scalable community-based rehabilitation models.
Presentation Type
Poster
Recommended Citation
Ganesh, Anjana and Baker, Kelsey, "Assessing the use of Marker-less Motion Capture Systems for Neurological Disorders in Indoor vs Outdoor Settings." (2025). Research Colloquium. 10.
https://scholarworks.utrgv.edu/colloquium/2025/posters/10
Included in
Assessing the use of Marker-less Motion Capture Systems for Neurological Disorders in Indoor vs Outdoor Settings.
Background: Marker-less motion capture systems have improved the assessment of human movement by allowing for a non-invasive, affordable, quantitative data collection useful in clinical settings. However, their use in real-world community care settings is largely underexplored, presenting a unique opportunity to bridge the gap between controlled laboratory environments and the complexity of providing care in diverse, everyday contexts. This systematic review examines the differential use of MMC systems across real-world indoor environments (clinics, homes, living labs) and outdoor settings (public walkways, community spaces), with a focus on their application to neurological disorders such as stroke and neurodegenerative diseases.
Methods: A systematic review was conducted using databases including PubMed, Web of Science, MEDLINE (NIH) and IEEE for studies published since 2010. Search terms utilized included "marker-less motion capture," "indoor," "outdoor" "gait" "neurodegeneration," and related phrases. The initial search yielded 63 articles. After removing duplicates, 41 articles remained for screening. Following a detailed review of titles and abstracts against predefined inclusion and exclusion criteria, 7 articles were selected for the final full-text analysis. The papers were analyzed to identify key theme of indoor versus outdoor applications.
Results: Indoor MMC systems demonstrated high reliability and ease of integration in home and clinical settings. Gait velocity measurements yielded ICCs between 0.88 and 0.94, and step length errors were less than 2.5 cm. Upper limb motion capture achieved joint angle ICCs of ~0.87. Usability of these systems were high for both patients and clinicians with scores exceeding 80/100. These systems are also able to be integrated with telehealth platforms, offering consistent data capture under controlled lighting and spatial conditions. In contrast, outdoor MMC systems provided the advantage of ecological validity but showed decreased precision. Joint tracking mean absolute errors ranged from 4.2 to 6.7 cm, and lighting variability, occlusions, and background interference contributed to a 15–25% reduction in data fidelity. Temporal alignment issues resulted in errors up to 100 ms.
Conclusion: Indoor MMC systems currently offer improved measurement accuracy, reliability, and user integration for neurorehabilitation applications, making them more feasible for clinical and home-based deployment. Outdoor systems, while technically limited, provide unique insights into real-world movement patterns that are essential for functional outcome assessments. Assessing the improvements of computer vision, AI-driven occlusion, and sensor fusion can help bridge the gap between the systems. Future research should focus on longitudinal validation of these systems with comparison to clinical metrics to improve equity and inform scalable community-based rehabilitation models.
