For care facilities

Fall coverage that passes a privacy review.

Cameras in resident rooms are usually a non-starter — for good reason. EMOTE4D is the version operators can actually defend: a sensor that never keeps an image, is built to avoid the false alarms that erode staff trust, and keeps working when the network doesn't.

The procurement question

Privacy that survives a procurement review.

In a regulated residential setting, the hardest question isn't “does it work” — it's “what happens to the footage.” Most camera systems answer that with a policy. EMOTE4D answers it with architecture.

Because no image is ever stored or transmitted, the answer to “where does the footage go?” is that there is none. The camera's frame becomes a stick figure on the device itself and is discarded in the same moment; nothing else leaves the room. There is no cloud account that can be breached, subpoenaed, or sold.

There's no facial recognition either — it reads posture, not identity. It's a shorter conversation with a privacy officer, and with residents' families.

  • No raw video stored or transmitted
  • No facial recognition — the system reads posture, not identity
  • Processing stays on the device, in the room
Where the data goes — and where it can't Flow diagram. A camera watches the room. On the device, each frame becomes a stick figure and the image is discarded on the spot. From the device, one path leads to an SMS sent to the responders you designate. A second path, toward cloud storage, is crossed out: there is no footage — nothing stored, nothing sent. The room A camera watches. On the device Each frame becomes a stick figure; the image is discarded on the spot. There is no footage. Nothing stored. Nothing sent. A text to staff To the responders you designate.
The one-glance answer for a privacy review: a stick figure on the device, a text out — and no footage path at all.

Fewer, truer alerts

Built against alert fatigue.

An alarm staff learn to ignore is worse than no alarm. When false alerts pile up, staff stop trusting all alerts — and the one that matters gets ignored.

EMOTE4D is designed around that problem. A fall has to clear two independent checks before anyone is notified: a model first screens the motion, then a separate step confirms the person is actually on the ground, with a short verification delay before the text goes out. The discipline is deliberate: fewer, truer alerts. The goal isn't to flag everything that might be a fall — it's to be worth paying attention to when it does alert.

Here is that sequence running, drawn from the same stick figure the device actually produces.

  1. Check one — screen. The motion looks like a fall.
  2. Check two — verify. The person really is on the ground.
  3. Short delay, on purpose. Compressed in this demo. It exists to cut false alarms.
SMS · to the caregivers you chose

Images stored during this event: 0  ·  Images sent: 0

No cloud dependency

The safety function doesn't depend on the cloud.

Detection runs entirely on the device in the room. It keeps working through a Wi-Fi outage or an internet provider problem, because knowing a fall happened never required a round-trip to a server in the first place.

There is no per-room cloud subscription required for the core safety function, and no monthly fee gating whether the system actually watches for falls.

Day to day

Designed for how care teams actually work.

Alerts where staff already are

A confirmed fall triggers an SMS to the responders you designate, with a short verification delay built in.

Simple to place

One device per room, positioned to cover the space — no wearables for residents to remember, charge, or refuse.

Quiet by default

No screens for residents, no two-way audio surveillance, no always-on video feed for anyone to watch.

These are design intentions for the pilot, not validated results. We'll have honest numbers to share once the pilot has produced them.

The pilot

What a pilot looks like.

We're looking for a small number of care settings in Manitoba to run EMOTE4D in real rooms, in 2026, and to tell us the truth about it.

A pilot means installing the device in a set of rooms, routing confirmed-fall alerts to the staff or workflow you choose, and sitting down with us regularly to review what fired, what didn't, and what needs to change. The point of a pilot is to find the gaps before anyone relies on it — so we'd rather hear what's wrong than what's polite.

We'll be measuring how the system performs in the conditions that actually matter — night, clutter, real movement — rather than leaning on lab numbers. And we're committed to independent testing and to reporting what we find, including the limitations.

Start a conversation

Pilot with us

Bring the pilot to your facility.

We're starting in Manitoba in 2026 and talking with operators now. Tell us about your facility and we'll come to you with specifics, not a sales deck.

Want to say more than an email address? Use the full form.

No spam, no drip campaign. We write when there's something real.