Validation of MoveLab smartphone gait and mobility assessment against gold-standard motion capture

Research question

Can gait and functional-mobility measures derived from a smartphone’s built-in inertial sensors, captured through a browser-based application, agree closely enough with established reference methods to be useful as objective movement measurement?

Gait measures were compared against marker-based 3D motion capture. Timed Up and Go and 30-second Sit-to-Stand were compared against clinical-standard manual scoring.

Population

Participants

25 healthy volunteers

Sex

14 female, 11 male

Mean age

31.8 ± 11.6 years

Gait history

No history of gait impairment

All participants were healthy adults recruited for laboratory testing.

Technology tested

A browser-based React data collection application developed by Agile Kinetic. It accessed the smartphone’s built-in inertial sensors at a sampling rate of at least 50 Hz.

Device

Samsung Galaxy A25

Placement

Pouch secured around the waist

Sensors

Built-in smartphone inertial sensors

Sampling rate

At least 50 Hz

Comparator / reference standard

• Gait: Qualisys marker-based 3D motion capture — the gold-standard laboratory reference for spatiotemporal gait measurement.
• Timed Up and Go: clinical-standard manual timing.
• 30-second Sit-to-Stand: clinical-standard repetition counting.

Protocol

1. 1Participants attended a movement-analysis laboratory instrumented with a Qualisys marker-based 3D motion capture system.
2. 2A smartphone running the browser-based MoveLab research application was secured in a pouch at the waist.
3. 3Walking trials were recorded simultaneously by the smartphone application and the motion capture system.
4. 4Timed Up and Go was performed and scored by the smartphone application alongside clinical-standard manual timing.
5. 530-second Sit-to-Stand was performed and counted by the smartphone application alongside clinical-standard repetition counting.
6. 6Smartphone-derived measures were compared with the corresponding reference measures using intraclass correlation coefficients (ICC).

Results

Agreement varied by parameter. Temporal gait measures — step time, cycle time, stance time and cadence — showed the strongest agreement with motion capture. Double-support measures were the weakest.

Agreement for gait speed, step length, stride length, swing time, cadence and gait timing ranged from moderate to good. Double support, initial double support and terminal double support fell below that range, indicating that these measures were less reliably reproduced from waist-mounted smartphone inertial data in this study.

Interpretation

Under controlled laboratory conditions, with the smartphone secured at the waist and in healthy adults, several gait and functional-mobility measures derived from browser-captured smartphone inertial data showed moderate-to-good agreement with reference methods.

Because agreement is parameter-specific, the measures should be considered individually rather than treating the whole assessment as uniformly accurate. Double-support-related parameters in particular require further development before they can be relied upon.

What this study validated

• That gait speed, step length, stride length, step, cycle, stance and swing time and cadence can be derived from waist-mounted smartphone inertial data with moderate-to-good agreement against motion capture in this population.
• That Timed Up and Go and 30-second Sit-to-Stand scores produced by the application agreed with clinical-standard scoring in this population.
• That a browser-based application can capture smartphone inertial data at a sufficient sampling rate for this analysis.

Conclusion

The study provides peer-reviewed evidence that smartphone inertial data captured through a browser-based MoveLab research application can generate several gait and functional-mobility measures with moderate-to-good agreement against reference methods in healthy adults under controlled laboratory conditions.

Full citation

Powell K, Amer A, Glavcheva-Laleva Z, Williams J, O’Flaherty Farrell C, Harwood F, Bishop P, Holt C. MoveLab®: Validation and Development of Novel Cross-Platform Gait and Mobility Assessments Using Gold Standard Motion Capture and Clinical Standard Assessment. Sensors. 2025;25(18):5706.

DOI: 10.3390/s25185706. Published 12 September 2025.