Hytro Movement Analysis Dissertation

Project Overview

This project was conducted in collaboration with Hytro, a company operating in the training and rehabilitation space with a specific focus on blood flow restriction (BFR) garments. While the physiological effects of BFR are well established and had already been extensively tested by the company, Hytro identified an unresolved engineering question relevant to product design and performance.

The core problem posed by Hytro was mechanical rather than physiological: does the structure of a blood flow restriction garment influence how people move during dynamic activities such as running and cycling? In other words, beyond altering blood flow, could the physical construction of the garment itself meaningfully affect movement mechanics?

The objective of this project was to experimentally evaluate whether different BFR garment designs alter lower-limb kinematics compared to standard athletic shorts and competitor products.

Experimental Design and Concept Development

To address this question, I designed a 3D motion capture study comparing multiple garment conditions, including Hytro’s products, competitor BFR garments, and standard shorts as a control condition.

Twelve trained runners and cyclists were recruited to complete short bouts of running and cycling under each garment condition. Each participant was instrumented with 28 reflective markers, and movement was captured using a 24-camera Qualisys motion capture system. This setup enabled high-resolution tracking of lower-limb kinematics during dynamic movement.

The experimental concept focused on isolating garment structure as the independent variable while keeping movement tasks, speed, and testing conditions consistent across trials.

Data Processing and Measurement Approach

Following data collection, a substantial portion of the project focused on data processing and analysis. An initial attempt to use an automated AIM model for marker labeling was abandoned due to poor performance and reliability. Instead, trials were manually cleaned and labeled, which significantly increased processing time but ensured higher data quality.

A custom Python-based analysis pipeline was then developed to:

• Fill marker gaps

• Estimate joint centers

• Filter trajectory data

• Construct local coordinate systems

• Calculate lower-limb joint angles across all planes of motion

To statistically evaluate differences between garment conditions, statistical parametric mapping (SPM) was applied to the kinematic time-series data, allowing for waveform-level comparisons rather than discrete point analysis.

Experimental Findings

Due to marker occlusion issues and the time-intensive nature of manual labeling, only the running data from six participants could be fully processed and analyzed.

Within this limited dataset:

• Only one statistically significant difference between garment conditions was identified

• Several small, non-significant trends were observed

• No consistent or meaningful changes in overall running kinematics were found between BFR garments and control conditions

These results suggest that, within the constraints of the analyzed data, the structure of both pneumatic and practical BFR garments does not substantially alter running mechanics.

Key Insights and Implications

The primary takeaway from this project is that BFR garment structure alone is unlikely to meaningfully influence gross lower-limb running kinematics. However, the findings must be interpreted cautiously due to the reduced sample size and data quality challenges.

Importantly, this project highlights that null or weak findings can still provide valuable insight for product development. The results suggest that concerns about unintended mechanical interference from garment design may be less critical than initially assumed, though further testing with improved data capture would be required to confirm this conclusively.

With a full participant dataset and cleaner motion capture trials, future work could provide a clearer understanding of whether subtle kinematic or coordination changes emerge under different garment designs or activity intensities.

Skills Demonstrated

• Biomechanics and human movement analysis

• 3D motion capture experimental design

• Advanced data processing in Python

• Statistical parametric mapping (SPM)

• Experimental problem-solving under real-world constraints

• Translating industry questions into biomechanical testing protocols