The science
How Step Doctor measures movement
Step Doctor turns a phone video into a 3D skeleton and measures gait, the Timed Up and Go and the SPPB from where the joints actually are. Every result is judged against published research, and every result can be checked against its 3D replay.
From phone video to measurement
Record
A short video with the phone's camera: a walk toward the phone, an iTUG, or the three SPPB tests.
Find the person
The person being tested is found and followed on every frame, even with other people in view.
Reconstruct in 3D
30 body points per frame, in metric 3D: feet, knees, hips, trunk, arms and head.
Find the events
Heel strikes for gait; the six iTUG phases; for the SPPB, each balance hold, the timed walk and every chair rise.
Report in context
Each measure is graded against published norms and shown with a 3D replay of the test.
A camera, compared with a phone in a pocket
Smartphone gait tools mostly use either the phone's motion sensor, carried on the body, or its camera. Both have strengths. Step Doctor is built for standardized tests, where seeing the movement matters most.
| Question | Step Doctor (camera, 3D pose) | Phone-in-pocket motion sensing |
|---|---|---|
| What is measured | Joint positions directly in 3D: feet, knees, hips, trunk, arms | Acceleration and rotation of one point near the hip; gait events are inferred |
| Can you see what was measured? | Yes. The 3D replay shows every step and phase | No. The signal isn't human-readable |
| Placement | Nothing on the body. Keep the person in frame | A pocket or a strap |
| Whose phone? | Any phone. A clinician or a family member can film | Usually the person's own phone, carried on the body |
| Whole-body posture | Trunk lean, arm swing, knee bend and standing height | Limited to what one body point reveals |
| iTUG and SPPB | From posture: hip height, heading and distance show every phase, hold and rise | Inferred from changes in acceleration |
| All-day monitoring | Not the goal: Step Doctor measures standardized tests on demand | A strength of body-worn sensing |
Judged against published evidence
Gait speed, TUG time and the SPPB are among the most studied predictors of falls and decline in older adults. Step Doctor reports each against the thresholds from that research.
| Measure | Threshold or effect | Evidence |
|---|---|---|
| Gait speed | Each 0.1 m/s slower: about 7% higher risk of falling | Verghese et al., 2009 |
| Gait speed | Normal adult walking speed is about 1.2–1.4 m/s; an improvement of 0.1 m/s or more is useful | Fritz & Lusardi, 2009 |
| Gait speed | Under 0.6 m/s (a cut-off from earlier studies): more likely poor health and function; about 0.8 m/s: average life expectancy; 1.0 m/s or more: longer survival than expected | Studenski et al., 2011 |
| Gait speed | Each 0.1 m/s faster: lower mortality, in 34,485 older adults | Studenski et al., 2011 |
| Gait variability | Higher stride-time variability predicted falls over one year | Hausdorff et al., 2001 |
| TUG time | Average times in healthy older adults: 8.1 s at 60–69, 9.2 s at 70–79, 11.3 s at 80–99 | Bohannon, 2006 |
| TUG time | 13.5 s or more identified fallers with 87% sensitivity and specificity in a small study | Shumway-Cook et al., 2000 |
| TUG time | Pooled data: a time of 13.5 s or more makes falls more likely, but a faster time doesn't rule them out (sensitivity 31%, specificity 74%); not to be used alone | Barry et al., 2014 |
| iTUG phases | Reference times for each phase in older adults without cognitive impairment; turn-to-walk was slower in mild cognitive impairment. Step Doctor grades a phase at the study's mean + 1 / + 2 SD | Mirelman et al., 2014 |
| SPPB total | 0–12, bands 10 / 7 / 4: lower scores predicted mortality and nursing-home admission | Guralnik et al., 1994 |
| SPPB cut-offs | Points per test for the walk and the chair stands (protocol figure) | Rodríguez-Mañas et al., 2014 |
| SPPB change | About 0.5 point is a small meaningful change; about 1 point a substantial one | Perera et al., 2006 |
| SPPB change | Meaningful change estimated in an older-adult trial (LIFE-P) | Kwon et al., 2009 |
In numbers
- 3
- standard tests: gait, iTUG and SPPB
- 30
- body points tracked in 3D on every frame
- 6
- iTUG phases, each timed and graded
- ~4.5 s
- median analysis time after upload (496 analyses, October 2026)
What has been validated
The thresholds above come from peer-reviewed studies of the tests themselves. Step Doctor's own measurements were compared with a gold-standard pressure-sensor gait mat in an internal study: 93% clinical accuracy. That study is not yet peer-reviewed; we will publish its method and results.
Step Doctor supports screening and monitoring. It doesn't diagnose, and results should inform, not replace, the judgement of a healthcare professional.
Quality and data protection
AICU's quality management system is certified to ISO 13485:2016 (certificate KM 26146, Kiwa Korea).
Videos and results are encrypted in transit and at rest. Videos are used only to produce your results and are deleted from our servers after about 72 hours. Results are never sold or used for advertising.
References
- Barry E, et al. Is the Timed Up and Go test a useful predictor of risk of falls in community dwelling older adults: a systematic review and meta-analysis. BMC Geriatr. 2014;14:14.
- Bohannon RW. Reference values for the Timed Up and Go test: a descriptive meta-analysis. J Geriatr Phys Ther. 2006;29(2):64–68.
- Fritz S, Lusardi M. White paper: "Walking speed: the sixth vital sign." J Geriatr Phys Ther. 2009;32(2):2–5.
- Guralnik JM, et al. A short physical performance battery assessing lower extremity function. J Gerontol. 1994;49(2):M85–M94.
- Hausdorff JM, Rios DA, Edelberg HK. Gait variability and fall risk in community-living older adults. Arch Phys Med Rehabil. 2001;82(8):1050–1056.
- Kwon S, et al. What is a meaningful change in physical performance? J Nutr Health Aging. 2009;13(6):538–544.
- Mirelman A, et al. Association between performance on Timed Up and Go subtasks and mild cognitive impairment. J Am Geriatr Soc. 2014;62(4):673–678.
- Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54(5):743–749.
- Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39(2):142–148.
- Rodríguez-Mañas L, et al. An evaluation of the effectiveness of a multi-modal intervention in frail and pre-frail older people with type 2 diabetes - the MID-Frail study: study protocol for a randomised controlled trial. Trials. 2014;15:34.
- Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls in community-dwelling older adults using the Timed Up & Go Test. Phys Ther. 2000;80(9):896–903.
- Studenski S, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50–58.
- Verghese J, Holtzer R, Lipton RB, Wang C. Quantitative gait markers and incident fall risk in older adults. J Gerontol A Biol Sci Med Sci. 2009;64(8):896–901.