Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Master of Science

Program

Physical Therapy

Collaborative Specialization

Machine Learning in Health and Biomedical Sciences

Supervisor

MacDermid, Joy C.,

Abstract

This thesis applies Phase Space Dynamics (PSD) to enhance shoulder biomechanics analysis during the FIT-HaNSA test. Analyzing videos of seven healthy individuals from the HULC database, shoulder movements were recorded during a 5-minute task. Kinematic data, extracted using Dartfish® software and processed in Python, were used to calculate eleven PSD features reflecting movement variability and efficiency. Results demonstrated distinct patterns in shoulder movement dynamics, with high concordance between PSD features and key indicators like fatigue, compensatory movements, and increased intensity. Notably, PSD provided a more comprehensive and objective analysis compared to traditional methods. The study concludes that PSD is a promising tool for advancing shoulder biomechanics, offering valuable insights for physiotherapy and rehabilitation.

Summary for Lay Audience

This thesis explores an innovative method called Phase Space Dynamics (PSD) to enhance the analysis of shoulder movements, particularly during the FIT-HaNSA test, a common assessment of shoulder function involving the repetitive transfer of objects between shelves. Traditional techniques often fail to capture the intricate and complex nature of shoulder movements, but PSD offers a more detailed and accurate analysis.

Using videos of seven healthy individuals performing the FIT-HaNSA test, shoulder movements were recorded and analyzed. Kinematic data was extracted using Dartfish® software, and then processed in Python to calculate eleven PSD features reflecting movement variability and efficiency. The analysis showed that PSD could identify subtle differences in shoulder movement patterns, providing important insights into shoulder function. For instance, it was able to detect signs of fatigue, compensatory movements (when the body adjusts to reduce strain), and increased movement intensity more accurately than traditional video analysis.

In one part of the study, PSD revealed distinct patterns in shoulder movement dynamics between different participants. For example, Participant 1 exhibited higher variability, complexity, and dispersion in movement patterns, suggesting more flexible or adaptive strategies compared to Participant 3. Another part of the study focused on the concordance between PSD features and key indicators of shoulder performance, such as vocal expressions of fatigue, compensatory movements, and increased movement intensity. The results showed a high level of agreement, highlighting PSD's ability to provide a more comprehensive and objective understanding of shoulder biomechanics.

The implications of these findings are significant for clinical practice. By capturing detailed patterns that traditional methods often miss, PSD enhances clinical assessments and can lead to better-targeted rehabilitation methods. This, in turn, can improve the recovery process for individuals with shoulder injuries. Overall, this thesis highlights the potential of PSD as a powerful tool for advancing shoulder biomechanics. By providing a deeper and more precise analysis of shoulder movements during everyday tasks, PSD contributes valuable insights for physiotherapy and rehabilitation, ultimately leading to better diagnosis and treatment strategies for shoulder conditions. This new approach promises to improve the way we understand, diagnose, and treat shoulder-related issues

Download

Share

COinS