A capability locked in the lab
Detailed movement analysis has existed for decades, but it lived in specialized facilities — research centers and a small number of academic hospitals with the cameras, markers, and staff to run it. The science was mature; the access was not. For the vast majority of clinicians and patients, lab-grade movement measurement simply wasn't on the table.
What made it portable
Two shifts changed the equation. First, computer-vision models grew accurate enough to identify body landmarks directly from ordinary video, removing the need for markers and specialized cameras. Deep-learning pose estimation reached human-comparable tracking with relatively little labeled training data (Mathis et al., 2018), and models working from single-camera clinical video went on to predict gait measures such as walking speed and cadence — an advance its authors described as increasing access to quantitative motion analysis in clinics and at home (Kidziński et al., 2020). Second, the cameras themselves — the ones in everyday smartphones — became powerful enough to capture the necessary detail. Together, these advances moved the heavy lifting from dedicated hardware into software.
Why the location matters
Bringing measurement to the point of care isn't just convenient; it changes what's measurable. Assessments can happen where the patient actually is — the clinic, the bedside, the home — and they can happen often enough to reveal trends rather than isolated snapshots. Frequency and context are exactly what episodic, lab-bound testing could never provide.
Keeping rigor in the move
Portability only matters if the measurement stays trustworthy. Moving out of the lab means being disciplined about standardized capture and honest about where accuracy is strongest — current reviews find markerless spatiotemporal gait measures broadly comparable to marker-based capture, with joint angles still needing more caution (Wade et al., 2022). Our validation page summarizes the evidence. Used within those bounds — and alongside clinical judgment rather than in place of it — portable movement analysis extends a once-rarefied capability to the everyday settings where most care is delivered.
References
- Mathis A, Mamidanna P, Cury KM, et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nature Neuroscience. 2018;21(9):1281-1289. doi:10.1038/s41593-018-0209-y
- Kidziński Ł, Yang B, Hicks JL, Rajagopal A, Delp SL, Schwartz MH. Deep neural networks enable quantitative movement analysis using single-camera videos. Nature Communications. 2020;11:4054. doi:10.1038/s41467-020-17807-z
- Wade L, Needham L, McGuigan P, Bilzon J. Applications and limitations of current markerless motion capture methods for clinical gait biomechanics. PeerJ. 2022;10:e12995. doi:10.7717/peerj.12995