Clinical question
Can consumer smartphone videos estimate body kinematics and dynamics without a conventional motion-capture laboratory?
Study design
The Stanford research team combined pose estimation, biomechanical modeling and physics-based simulation. In a laboratory validation with ten healthy participants, measurements from two iPhones were compared against marker-based motion capture and force plates.
What the researchers found
OpenCap estimated joint angles with a mean absolute error of 4.5° across the evaluated activities. The work also assessed dynamic outputs such as joint moments and ground-reaction forces.
Why this matters for MedMeters
The study motivates MedMeters’ ambition to make reproducible movement measurements accessible outside specialized motion labs. It also suggests a sensible validation strategy: compare each protocol against an appropriate reference.
What this does not prove
This is a two-phone system in a small healthy cohort, not validation of a single-phone MedMeters protocol. Joint-specific errors and protocol context matter; an error of several degrees may exceed the smallest change that a clinician hopes to detect.
OpenCap: Human movement dynamics from smartphone videos
Uhlrich et al. · PLOS Computational Biology · 2023
DOI: 10.1371/journal.pcbi.1011462