Surgical phase
under development
Where the case is, decided while the case is running. Anaesthesia prep, surgical prep, surgery, wrap-up.
Phase is the spine of every other readout. Turnover intervals, sterile-phase traffic and care-bundle timing all depend on the team agreeing where the case started and stopped.
The decision is made live and never revised, which is harder than labelling a recording afterwards and is the only version that is useful to a team in the room.
Decided once per second while the case runs, and developed against a public dataset of simulated surgery on a phantom. No clinical footage yet.
Sterile technique
open research
Contact between sterile and non-sterile: gown and glove breaches, field contact, instrument handling, gowning sequence.
This is the hardest of the five and we are not claiming it works yet. A better model of the room does not automatically catch more breaches, and our own testing has not shown that it does.
A monitor that flags constantly is worse than no monitor, so we count false alarms with the same weight as misses.
A breach and the phase it happened in come from the same description of the room, so the two can never contradict each other.
Traffic and turnover
open research
Door events and person counts during sterile phases, and the interval between one case closing and the next opening.
Turnover is the number theatre managers already fight over, and it only means anything if the end of one case and the start of the next are called the same way every time.
Nothing counts people separately. The same description of the room that gives you phase gives you who was in it.
Every additional entry during a sterile phase degrades operating-room air, which is the documented mechanism linking traffic to infection risk.
Care-bundle events
open research
The prevention checklist as timestamped events: skin prep, gowning and gloving, timeout, count, sterile-field discipline.
Prevention bundles are checklists that hospitals are audited on and that nobody can actually measure. Making them measurable is the reason the company exists.
The events are read from video. Nothing is typed in by a member of staff, and nothing is reconstructed from the notes afterwards.
The checklist comes from our own clinical team and follows the prevention bundle a theatre already works to. No outcome data yet, on any site.
Metric 3D twin
open research
A metric reconstruction of the room from multi-view RGB alone, in which every point carries an entity label.
Most sterile-technique rules are really questions about distance: how close, how far, did it touch. Those cannot be answered from flat video.
It also means a ceiling camera and no other hardware. Depth sensors in a working theatre are a procurement and sterilisation problem before they are a technical one.
Cameras only, with no depth sensor in the room. Distances are checked against a laser measurement of the same scene.
