A video based method to quantify posture of the head and trunk in sitting

Area of Science:

• Biomedical Engineering
• Clinical Biomechanics
• Rehabilitation Science

Background:

• Maintaining an upright posture in sitting is crucial for neuromotor control assessment.
• Current clinical posture assessments are often subjective and lack quantitative precision.
• Existing technologies for posture quantification are difficult to implement in clinical settings.

Purpose of the Study:

• To develop and validate a video-based method for numerically measuring postural alignment of the head and trunk in sitting.
• To assess the reliability and agreement of this novel method compared to a 3D motion capture system.

Main Methods:

• A video-based system and a 3D motion capture system simultaneously recorded static and dynamic sitting postures of 12 healthy adults.
• Segmental angles of the head, neck, and trunk were calculated.
• Clinically experienced experts defined aligned posture frames, and agreement was assessed using intraclass correlation (ICC).

Main Results:

• Expert agreement on aligned posture definition was excellent (ICC=0.99).
• The video-based method showed strong agreement with the 3D system (below 3.8° for static, 8.4° for dynamic trials).
• The method provided detailed trunk/spinal profiles, surpassing previous techniques.

Conclusions:

• The developed video-based method reliably quantifies sitting posture alignment.
• This technique offers a practical, quantitative tool for clinical use, complementing subjective assessments.
• It has potential applications for evaluating patients with diverse pathologies affecting posture.