medmeters.
EXTERNAL RESEARCH · GAIT & COMPUTER VISION

Can ordinary cameras support the Functional Gait Assessment?

A 2026 feasibility study tested whether three cameras and markerless pose estimation could extract clinically useful gait measurements during a familiar ten-task assessment.

Published June 2, 2026 · Editorial summary prepared October 2026 · Independent research, not MedMeters clinical validation.

Illustrative camera-based gait assessment in a rehabilitation clinic
MedMeters editorial review prepared October 2026. June 2026 is the original study's publication date. This article is not a clinical validation of MedMeters.
R² 0.95–0.98Correlation-derived determination values for selected gait parameters in a tiny feasibility cohort

Clinical question

Can ordinary cameras and markerless pose estimation measure aspects of walking that are relevant to the standardized Functional Gait Assessment (FGA)?

Study design

Osman Darici, Chanel Cabak and Jeremy D. Wong developed a three-camera markerless motion-capture setup and compared its measurements against traditional marker-based motion capture in three participants. They also showed feasibility in five older adults performing FGA tasks.

What the researchers found

Reported R² values were 0.98 for step length, 0.97 for step width and 0.95 for head speed compared with the reference system. This supports technical feasibility under the tested conditions, not broad clinical accuracy claims.

Why this matters for MedMeters

FGA has ten clinician-rated tasks, many involving speed control, step placement, head movements and balance. Supplementary camera-derived data could make repeat measurements easier to review and document. Time savings must still be measured from camera setup all the way to the final EHR record.

What this does not prove

The validation cohort was N=3, and the demonstration involved N=5. The study used three cameras. It does not prove that a single smartphone is adequate, that automatic ordinal FGA item scoring is valid, or that clinicians save time. Those would require separate studies and validation.

Original peer-reviewed publication

Darici O, Cabak C, Wong JD (2026). A low-cost markerless motion capture system to automate functional gait assessment: Feasibility study. PLOS ONE 21(6):e0346606.
DOI: 10.1371/journal.pone.0346606 ↗

External scientific evidence summary. Not medical advice. MedMeters is a research-stage product concept, not evaluated in this study. Medical-device classification, performance and regulatory claims require separate verification.