Healthcare · Strong theory

Bed-exit and unassisted-egress prediction for high fall-risk patients

Warns staff seconds before an unsteady patient tries to get out of bed alone.

Strong theory Strong theory live in runtime

The physics is solid; our own validation is scheduled.

Before someone stands up, they sit up, swing their legs over, and shift their weight — a little dance the WiFi waves can see. The system spots that pattern and warns staff before the feet hit the floor, not after a fall.

In the physics: The patient's dominant reflection centroid translates from the horizontal bed plane toward a vertical seated/standing posture and the RTI energy migrates from bed cell to floor/edge cells; postural micro-Doppler signature of weight-shift precedes egress.
Preparing the recorded fieldLoading real LatentField pipeline output…

This is real pipeline output for Healthcare, replayed. The exact reading depends on your room.

Bring this into the real roomPatient Room Field Node3 sensing nodes · ~$140 parts · about 60 min
Go deeper

How the field becomes an answer

RTI differential imaging tracks body centroid vs. learned bed footprint Doppler posture-transition features compartment Kalman dynamics predicts trajectory threshold on egress-intent state

How we prove it

Presence/occupancy (r=0.9) and localization proven; validate on WiMANS multi-position activities + own-hardware bed-egress scripted trials against a pressure-sensitive bed mat.

Who it is for

'Predictive bed-exit alerting' — replaces high-false-alarm pressure pads for fall-prevention program managers.

runtime Runs today in LatentField 2.0 as fall_detect

Complete first-room estimatePatient Room Field Node3 nodes · ~$140 parts · 60 min Request kit, subscribe, or set up →
Compatible DIY hardware options
Practitioner Discrete$60–90

15 min · beginner

Operator Single-Venue$280–380

120 min · advanced

← All Healthcare capabilities