One person is one person, no matter which system they walked into.
Registration, lab, radiology and billing each think they know who the patient is. Identity resolution is the service that decides, and it is unforgiving work — a wrong match is a clinical safety event, and a missed match is a duplicate chart nobody trusts.
Interfacing, identity, and imaging.
- HL7
Interfaces that stay up
ADT, ORM, ORU and the rest, over MLLP or files, between systems whose vendors have no interest in talking to each other.
- MPI
Master Patient Index
Deterministic and probabilistic matching across source systems, with confidence scoring, conflict surfacing and a reviewable trail for every merge.
- Imaging
Documents into the record
Scanned paper and faxes indexed and attached to the right encounter, so the chart is complete without anyone filing it by hand.
We built the Master Patient Index for Nassau University Medical Center. It went live roughly fifteen years ago and has run without an incident since. Software runs on infrastructure the client owns — we do not host clinical systems and we do not hold patient records on our own machines.
In production · fifteen years
Revenue cycle runs on the same discipline. For Episcopal Health Services and Hudson River Healthcare we drove software robotics on a terminal server — keying and reconciling across the clinical systems and the HIS the way staff do, screen by screen. At Queens Hospital we posted remittances at scale, read straight off the EOBs with intelligent character recognition — the ancestor of today’s document AI. All of it to keep the billing moving without a single interface the vendors were ever going to build.
Revenue cycle · software robotics against clinical systems
Tell us what you have been told is impossible.
A couple of sentences about the system and what you need out of it is enough to get a straight answer about whether we can do it.