Every step, read by AI.
The AI does the part nobody has time for: finding every step in a recording, measuring it, comparing it with what came before and writing it up — so you can decide what it means.
From raw signal to a report
- 01
Record
The phone on the thigh records how the leg moves — a raw, noisy signal no one could read by eye.
- 02
Find every step
The AI cleans the signal and marks each heel contact, right and left, across the whole walk.
- 03
Line the strides up
Every stride is laid over the others. When they overlap closely, the walking rhythm is steady.
- 04
Compare
The walk is set against the patient's previous visit, so even small changes stand out.
- 05
Report
Findings are written up in plain language, for the clinician to review and discuss.
Change the walk, watch the findings change
Pick a walk or move the sliders. The footprints, the overlaid strides and the findings respond the way a report would.
What the report would say
Walking pace
Comfortable
1.26 m/s
Rhythm
Regular
2.3% variation
Left / right
Even
98% symmetry
Illustration only: a simplified model to show how findings respond to a walk. It is not the StepAnalyser algorithm and the cut-offs are not clinical thresholds.
The measures behind every report
Familiar gait parameters, measured on every step of the walk instead of estimated from a glance down the corridor.
Walking speed
How quickly the patient covers ground — one of the most telling numbers a walk can give.
Cadence
Steps per minute, and whether that pace holds through the whole walk.
Left / right symmetry
Whether both legs take steps of the same length and timing.
Rhythm regularity
How closely one stride repeats the last. An irregular rhythm is easy to miss by eye.
Ground contact
How long each foot stays on the floor, and how long it spends swinging.
Change over time
Each result is set against the patient's own earlier walks, not only a textbook norm.
Compared with the patient, not just a norm
A norm is a wide netPopulation ranges for walking speed have to cover everyone of a similar age — so they are broad.
The AI highlights. You decide.
Decision support, not diagnosis
StepAnalyser points out patterns in the walk. Diagnosis and treatment decisions stay with the clinician.
Written for people
Findings come in plain language you can read in seconds and explain to the patient in the same appointment.
Built to be checked
Every report comes from a recording you can go back to — and compare with the next one.
Who the patient is stays apart from how they walk
- Consent is recorded before any walk is captured.
- Names and record numbers are kept apart from the walking data.
- Anonymisation options for research and exports.
See it with your own patients
Book a demo and we'll walk you through the app and an AI gait report, using the kind of patients you see every week.
Book a demo