One real client. Every call made by the clinician.
We ran Cliro on one real client’s records, with identifying details removed, inside the working rehabilitation practice we’re building in. It was the first of multiple real cases now under way at Engage VR Rehab — a multidisciplinary rehabilitation clinic in Newcastle, Australia, and the first in the country to bring virtual reality into rehab. This page says exactly what happened on that first case — and exactly what it does and doesn’t prove.
Cliro proposed the same first assessment the treating clinician chose — and flagged a treatable voice problem that usually goes unreferred. The clinician made every call.
n=1 / real de-identified client / the clinician made every call
How it was run. The records were de-identified before Cliro saw them. Cliro’s suggestions entered as proposals only — nothing became part of the record until the treating clinician looked at each one and decided. That gate isn’t a policy; it’s how the system is wired.
Why it matters. Agreeing with the clinician’s first assessment shows the reasoning lands where an experienced clinician lands. Flagging a treatable problem that usually goes unreferred is the more interesting half: the kind of catch that changes what care a person actually receives.
Recommended first assessment — matched to this client's age, diagnosis, goals, and screening results.
- 08:41:02 · cliro · proposed TUG · cited
What you're seeing: Cliro suggested a standard walking test for this client, and showed the research it comes from. Nothing happens until a clinician reviews it and agrees — try the confirm step yourself.
The engine works end-to-end on a real clinical case, inside a real practice, with the clinician-gate holding the whole way.
Generalisation. One client is one client — further real cases are under way, and university validation studies are in preparation, with institutions named once agreements are signed.
We publish the small number because it’s the true one. As the cases now under way conclude, this page grows with them.
This case study is a real client, so we show the outcome, not the record. To watch the engine work step by step — every event, every check, even a refusal — see Elena’s journey: a made-up client, run through the same real engine, where nothing is private because no one is real.