logo

Lab analyser connectivity

A new analyser is supported in 15 minutes, without touching your PC

Your analysers already know the result.
Somebody is still typing it in.

LISBridge is a small desktop application that sits between the machines in your lab and the software you run the lab on. The analyser finishes, the result appears on the patient’s report. It works with any hospital system or LIMS, it keeps working when the internet does not, and we supply and support it for you.

LISBridge running on desktop, laptop, tablet and phone
36
Analysers in the picker
40 models recognised by name
17
Models it can send orders to
Commissioned per site
15 min
Until a new analyser is supported
No reinstall, no visit
3
Ways to reach a machine
Serial, TCP, file pickup

What it replaces

A technologist reading a number off one screen and typing it into another.

It is the most ordinary error in a diagnostic lab and the hardest one to catch, because the number on the report looks exactly as authoritative whether it was measured or mistyped. A transposed digit in a creatinine is not visible to anyone downstream.

LISBridge removes the keyboard from that step. It installs on an ordinary PC in the lab, sits in the system tray, and listens to the analysers on the bench. When a machine finishes a sample, the result is read, checked, and handed to your hospital system against the right patient. Nobody retypes anything, and the report is ready sooner because nobody is waiting for a free moment to do the typing.

The analyser finishes and the result is already on the report. That is the whole product.

What you get
Runs on
Windows, macOS, Linux
Sits in
The system tray
Needs the internet
No
Analysers in the picker
36
Protocols
ASTM E1394, HL7 v2
Read from the source on 12 September 2026, not from a brochure.

How it is put together

One small application on one PC in the lab.

There is no appliance to rack and no box to buy. The analysers reach the lab PC over the cable they already have, or over the network. LISBridge reads what they send and forwards it to your server over plain HTTP, on your own network.

An analyser announces itself the first time it sends anything, so the list of machines in the lab builds itself rather than being typed in. If several PCs in a large lab need to do this at once, they can share one database and behave as a single installation.

Analyser → the lab PC → your server → the report
Sysmex XN-550AnalyserRS-232 or TCPLISBridgeon a PC in the lab36 named analysersparser updates itselfover the networkYour serveron the hospital LANPathology reportHb13.4 g/dLWBC7.2 ×10⁹/LPLT245 ×10⁹/LCreat.0.9 mg/dLresultover HTTPorders out — where the machine supports itQuality-control runs stay in the lab; only patient results are forwarded.Also reads standard ASTM and HL7 analysers beyond the named list.

The part worth understanding

Support for a new machine arrives over the network, not in a new version.

Most lab middleware hard-codes the machines it understands, so a new analyser means a new release of the software, a download, an installation, and usually a visit. Labs learn to expect weeks.

LISBridge keeps the knowledge of each machine in a separate sandboxed module that it fetches over the network and checks for every fifteen minutes. Adding a machine to the fleet is a parser release, not an application release. The installation in your lab picks it up on its own and the new analyser simply appears in the list.

Two useful consequences. The bytes coming off an analyser are read inside a sandbox rather than in the application itself, so a malformed message cannot take the bridge down. And the blood-count histograms are drawn inside that same module, so the curves improve when the parser does.

When you buy a new analyser
A parser is written
For that model
It reaches your PC in
15 minutes
You install
Nothing
You restart
Nothing
An engineer visits
No
The instrument list in the application updates itself along with the parser.

When the line drops

A lab that stops working because the internet did is not a lab.

The licence is signed and checked on the machine itself, against a key the application already holds. There is no call home to authorise a day’s work. A site with no connection at all keeps bridging results between the bench and its own server indefinitely.

The one thing that needs the internet is activating a new installation, and that is a single round trip. Afterwards, the checks that do run fail quietly: a missed licence check or a missed parser update changes nothing about the lab’s day.

With no internet at all
Analysers keep reporting
Yes
Results reach your server
Yes
The licence keeps validating
Yes
New analyser parsers
Wait for the line
Activation
Needs one connection, once

Quality control

Control runs are recognised as control runs, and never reach a patient.

Analysers do not announce which of their messages are clinical. Morning controls, calibrations and patient samples come down the same wire in the same format, and middleware that cannot tell them apart either floods the hospital system with control values or relies on somebody configuring rules correctly.

LISBridge classifies each message as it arrives. Control and calibration runs are recorded on their own screen for the lab to review, and are never written to a patient record or passed to your hospital system. A batch containing both is split correctly rather than rejected.

To be exact about the limit: the quality-control screen is a record of runs with control, lot, level, target mean and standard deviation. It is not a statistics package, and there are no Levey-Jennings charts or Westgard rules in it.

Every inbound message is sorted
A patient result
Goes to the report
A QC run
Stays in the lab
A calibration
Stays in the lab
A query from a machine
Answered, not stored as a result
Something unreadable
Kept raw, and flagged
A message that cannot be read is never guessed at — it is stored as it arrived and reported as a fault.

Reaching the machine

Three ways in, because analysers have not agreed on one.

Older instruments speak down a serial cable, newer ones over the network, and a few will only write a file to a folder. All three are ordinary here.

  • Serial, RS-232

    The cable already coming out of the back of the analyser. Pick the port and the baud rate and it listens.

  • TCP, either direction

    It can wait on a port for the analyser to connect, or dial out to the machine and reconnect by itself if the link drops.

  • Files over FTP, FTPS or SFTP

    For analysers that write a file instead of talking. It watches the folder and picks each result up, as often as every five seconds.

The other direction

Telling the analyser what to run, instead of somebody keying it at the machine.

Reading results is the mature half of this product and it is safe to rely on. Sending orders the other way — so the analyser knows which tests to run on the tube in front of it — is real, but it is not the same maturity, and we would rather say so than discover it at your bench.

Where an analyser can be told directly, it is. Where it can only ask, the order waits until the machine scans the barcode and asks for it. It is never both at once, so a sample cannot be run twice. Failed sends are retried in a proper queue and every attempt is on the record.

Of the seventeen models it can build orders for, one is validated against a proven working implementation and the rest are written from the manufacturer’s own specification and marked for validation before clinical use. That is why it ships switched off and is commissioned per site.

Orders out, honestly
Models it can build orders for
17
Validated against a working reference
1
Built from the maker's spec
The other 16
Switched on by default
No
Commissioned at your lab first
Always
We will tell you which of your machines is in which group before you buy anything.

What the operator sees

Eleven screens, and most days nobody opens any of them.

Logs rotate at 10 MB and keep five files, so the evidence for an incident is still there when somebody goes looking a week later.

Where we come in

You never open an account with a software vendor.

Imposetech is a LISBridge partner. In practice that means the licence is provisioned for you, the connection to your analysers is set up by our engineers, and when something stops you call the same number you already call. You do not negotiate a subscription, manage a renewal, or explain your problem to somebody who has never seen your lab.

It is not tied to our own hospital software. It connects to whatever you run — and if you do run MediSpa, lab connectivity is simply part of the same conversation instead of a second project with a second supplier.

Who does what
Licence
We supply it
Installation and setup
We do it
A new analyser
We arrange the parser
Support call
Comes to us
Your hospital system
Any HIS or LIMS

Straight answers

The questions labs ask

Will it work with my analyser?
There are 36 analysers in the picker and 40 models the parser recognises by name — Sysmex, Mindray, Roche Cobas, Horiba, Siemens, Abbott, bioMérieux, Snibe, Erba, Maccura, Diatron and others. Send us your model numbers and we will answer against that list rather than guessing.
What if my analyser is not on the list?
A parser gets written for it and delivered over the network. Your installation picks it up within fifteen minutes — no new version to install, nothing to restart, and nobody has to visit the lab. That is the part of this product we would defend hardest.
Does it need the internet?
Not to work. Activation needs one connection, once. After that the licence validates on the machine itself, and results keep flowing between the bench and your server whether or not the line is up. The only things that want the internet afterwards are new analyser parsers and the licence check, and both fail quietly.
Can it send orders to the analyser, not just read results?
For 17 models, yes, and it is switched off until you ask for it. One of those is validated against a proven working implementation; the rest are built from the manufacturer's specification and are commissioned and validated at your lab before anything clinical depends on them. We will tell you which of your machines is which.
What happens if our server is down?
Results keep arriving and keep being stored on the lab PC. Anything that could not be delivered shows as pending on the data screen and is sent on from there when the server is back. Nothing is silently lost, and nothing is silently retried either — a person presses the button.
Where does patient data go?
It stays on the lab PC and goes to your own server. Nothing about a patient or a result leaves your network. The software does send a small status heartbeat to the vendor — which analysers are connected and how many results ran today — and that carries no patient data and no result content.
Is this only for MediSpa?
No. It connects to any hospital information system or LIMS, and it is sold on its own. It happens to be built into MediSpa as well, which is why our lab customers rarely have to think about it.
Who do we call when it stops?
Us. We are a LISBridge partner: we supply the licence, set the connection up, and look after it afterwards. You never open an account with the software vendor or negotiate a subscription.
How is it priced?
By the number of PCs it runs on and the number of analysers connected, so a two-machine lab is not paying for a twenty-machine one. Ask for a quote with your analyser list.

Next step

Send us your analyser list

Model numbers are enough. We will tell you which of your machines can report into your system today, which can be sent orders, and what it would cost.