Validation of an optimized algorithm to use Kinect in a non-structured environment for Sit-to-Stand analysis.

Area of Science:

• Biomechanical analysis
• Human motion capture
• Rehabilitation engineering

Background:

• The Sit-to-Stand (STS) test is crucial for evaluating functional mobility.
• Accurate kinematic measurement of STS is essential for clinical assessment.
• Existing motion capture systems are often expensive and complex for non-clinical use.

Purpose of the Study:

• To develop and validate a low-cost, accessible method for reliable kinematic analysis of the STS test.
• To enable non-experienced users to perform motion analysis in everyday environments.
• To assess the accuracy of a novel algorithm using an RGB-Depth sensor compared to gold-standard systems.

Main Methods:

• Utilized a low-cost Microsoft Kinect (RGB-Depth sensor) for motion data acquisition.
• Developed a custom algorithm to analyze human motion in the sagittal plane during STS.
• Compared the proposed algorithm's performance against two commercial algorithms and a stereophotogrammetric system.

Main Results:

• The proposed algorithm achieved average errors of approximately 4 degrees for trunk/leg and knee angles.
• Demonstrated comparable accuracy to established commercial algorithms.
• Showcased the potential for reliable kinematic measurement in non-structured environments.

Conclusions:

• The developed algorithm offers a reliable and low-cost solution for STS kinematic analysis.
• This technology can be adopted in ambulatory or domestic settings by non-expert users.
• Paves the way for broader applications of accessible motion analysis in healthcare and beyond.