Next-generation telemedicine

Care that begins the moment you step inside.

CabinOS turns smart medical cabins into an active partner in the consultation — pre-screening, triaging, and remembering every patient across visits.

Scroll
Inside the cabin

The room is part of the consultation.

A sick person is a poor historian — they forget, they minimise, they rehearse. The cabin reads face, voice and symptom before a single question is asked, so the patient arrives already understood rather than interrogated.

Every signal is traceable to the model that produced it. The clinician decides; the machine only ever hands them evidence.

One visit, end to end

Six steps, and the last one feeds the first.

Most telemedicine is a funnel — book, call, leave. CabinOS closes: what the cabin learns today becomes context for the next visit.

  1. 01

    Arrive

    The cabin greets the patient and takes consent before anything is recorded.

  2. 02

    Observe

    A single frame reads visible signs of distress. No video is kept.

  3. 03

    Interview

    Symptoms in the patient's own words, plus tone, stress and speech rate.

  4. 04

    Triage

    A weighted score places them in the queue by clinical need, not arrival time.

  5. 05

    Consult

    The doctor joins already holding the report — and a copilot that listens.

  6. 06

    Document

    A drafted SOAP note the clinician edits and signs. Never auto-finalised.

Then it remembers. The completed visit is embedded and stored, so the next time this patient walks in, the doctor sees what changed since last time — without asking them to remember it.

The intelligence layer

What the cabin actually reads.

Four signals, each from a real model, each shown to the clinician with its provenance attached.

Facial affect

Seven-class expression read from one captured frame, returned with a confidence score and a clinical note.

Vision transformer

Voice biomarkers

Valence, arousal, vocal tremor, speech rate and pain likelihood — the things a patient rarely reports accurately.

Speech emotion model + DSP

Clinical language

Symptoms, medications and conditions lifted from the transcript as structured terms a doctor can scan.

Medical NER

Cross-visit memory

Past visits retrieved by meaning, so “heart attack” finds “myocardial infarction”. Empty on a first visit — never invented.

Medical embeddings + vector search

Built to be trusted with the thing that matters most.

Medical software earns trust by what it refuses to do. These are constraints in the system, not promises in a policy.

Media is never stored
Frames and audio are analysed in memory and discarded. Only the derived signals persist.
The database enforces access
Row-level security scopes every read to the patient it belongs to — not application code.
Signed notes are immutable
Once a clinician signs a note it cannot be edited, by anyone, including us.
Absence is shown as absence
If a model is unavailable, the field stays empty and says so. Nothing is inferred to fill a gap.

See what a consultation looks like when the room is paying attention.