Laboratory automation that works in routine use.
I help laboratories assess automation projects, develop liquid-handling methods and improve workflows on equipment they already own. The work includes sample preparation, deck programming, barcodes, worklists and the checks needed before routine use.
My strongest platform experience is Hamilton STAR and VENUS: the main programmer of a hospital department’s five Hamilton decks from 2015, and Hamilton Nordic’s application specialist in 2016. I have also run Eppendorf epMotion in routine, for PCR setup. I have handled a Tecan but not run one in routine, and I say so because those are different levels of experience.
- Hamilton STAR · VENUS
- Eppendorf epMotion, in routine
- 5 Hamilton decks, 1 Eppendorf, one department
- Remote first · Sweden, Europe and the Asia-Pacific
3 situations
Three situations I am called into
- Considering an investment
You are weighing a liquid handler and want the decision made on your sample volumes, batch sizes, tube types and the worklist your LIMS can produce, by somebody who does not sell robots. The outcome is a specification you can hold a vendor to, or the finding that a changed batch size or a different tube does the job.
- A workflow on equipment you own
The deck is in the room, and a method has to be developed or moved onto it and kept running: sample preparation, barcode matching, the worklist in and the results out, and an acceptance test on your own samples before the first routine run.
- An underused or unreliable method
A method that runs but fails often, needs an experienced hand beside it, or never made it out of validation. I review the deck layout, the order of steps, what the method does when a tip is missing or a rack is loaded a hole out, and how the output is judged, then fix what keeps it from routine.
Existing deck
The whole-blood case
The whole-blood preparation at Karolinska University Hospital, five people by hand next to a Hamilton STARlet the department already owned, for immunosuppressants in transplant patients and for phosphatidylethanol as an alcohol marker, is on the home page in brief; this is the method. I measured how long a sample stays mixed after it comes off the rocker, and whether a pipette entering below the plasma layer reports the same concentration as the mixed sample. It does, and the window is wide enough to work inside. The method was built around that window: agitate off the deck, de-cap, load in printed adapters so the tubes go on as they arrive, pipette wide-bore below the surface, and finish inside nine minutes.
- 5 → 1One operator preparing work that five people had prepared by hand
- ~½Turnaround, on the laboratory’s own figures
- 0Instruments purchased; the deck was already in the room
A full 96-well plate came out in 35 minutes with everything counted, equal or better than manual preparation on accuracy and precision, and the flow went into routine at 150 to 250 whole-blood samples a day. Those are results for one workflow in one high-volume hospital laboratory, with the deck already in the room; they show what a method looks like when the workflow is designed before the deck, not what yours will return. The deck work was presented as posters at IATDMCT congresses in Rotterdam in 2015 and Kyoto in 2017; the whole-blood flow is the Kyoto one.
Four working habits came with every deck I put a method on, and I bring them to yours:
- Barcode matching
The worklist and the tubes on the deck are matched barcode by barcode. A tube on the worklist but missing from the carrier is reported before the run; tubes loaded in any order are pipetted in the worklist’s order.
- Controls on the deck
The robot pipettes the calibrators, blanks and controls as well as the samples, so the same control that applies to a sample applies to the curve and the QCs.
- A manual twin
A manual version of every automated method is kept on par with it, with equal precision and bias demonstrated, so a breakdown changes the day’s pace and nothing else.
- Judged by the output
A programmed method is judged by what comes off the mass spectrometer, not by the deck reporting no errors. A plate-shaped heatmap of the internal standard response shows the systematic pipetting error that a step which ran without complaint hides.
That first method ran on a STARlet the therapeutic drug monitoring laboratory already owned. The department then bought a STAR M Plus for the doping control laboratory, which I programmed first as Hamilton’s application specialist and then as its user, two STAR Ms for drugs of abuse, another STAR for therapeutic drug monitoring, and an Eppendorf epMotion 5075 for pharmacogenetic PCR setups. Five Hamilton decks and an Eppendorf across one department, with me as its main programmer from 2015. The whole-blood flow with its figures is on the selected work page.
8 deliverables
What you get
Which of these an engagement includes depends on where you are. The first two are where every engagement starts.
Before the deck
- Workflow map
The process as it runs today, with where the time goes measured rather than reported.
- Suitability
Whether this workflow should be automated, what has to change first if so, and the alternative if not.
- Platform comparison
When you are choosing between platforms, I compare the options against your samples, volumes and operating requirements, explain the trade-offs and recommend a way forward, with the reasoning written down so you can disagree with it.
- Specification
What the automation has to do, in the laboratory’s terms and in terms a vendor can be held to.
On the deck
- Deck design
Deck layout, labware, carriers and adapters, the order of steps, and what the method does when a tip is missing or a rack is a hole out.
- Deck programming
The method itself in VENUS on a Hamilton, or on an Eppendorf epMotion, or a second pair of eyes on a method that exists. Programming, deck layout and workflow design are developed together.
- Integration and acceptance
The worklist from the LIMS, the barcodes the deck reads and matches to it, what goes back, and the join with the next instrument; then a written acceptance test the deck has to pass on your samples, with the pass criterion agreed before the run.
Into routine
- Procedures, training, fixtures
The procedure in your document system, the people trained on it, and the adapter, rack or holder that lets a sample go on the deck in the tube it arrived in. When the commercial part does not exist, I design and make it.
4 steps
How the work runs
- ReviewA day or two with the workflow, on site where it has to be seen, and a written proposal: which layer the delay is on, whether to automate, and what I would do first. The smallest piece of work that is useful to you.
- DesignThe workflow, the deck layout, the integration and the acceptance criteria, reviewed with the bench before anything is built.
- BuildThe method on the deck, the worklist join and the fixtures, with your staff rather than instead of them.
- Prove and hand overThe acceptance test on your samples against the agreed criterion; the procedure in your document system; the people trained; the deck running without me.
Most of this runs remotely: the mapping, the specification and the method itself can be written and reviewed from a desk, and site time is for what cannot. A longer on-site engagement is open for the right project. Engagements run from a couple of days to several months and are quoted in writing before they start; the commercial answers cover the rest.
Scope
Scope and responsibilities
I do not arrive with a robot to sell, and there is no instrument platform I represent. I am not a Hamilton reseller or service organization: spares, service contracts and the support line stay with the vendor. When an automated step changes the analytical method, the validation that follows is method validation, priced on its own. When the worklist, the registration or the instrument interface is where the problem lives, that is LIMS requirements, and the robot is downstream of it.
Six services
First step
Discuss your automation project
Describe the workflow in your own words: the sample types and roughly how many a day, where the queue forms, what equipment you have or are considering, and what you have tried. I will reply with the questions I need and tell you what I would look at first. The first conversation is free. The terms, travel and the other commercial answers are on the contact page. If you would rather not use a form, write to [email protected].
Contracting
- Consulting delivered through Ebihara Solutions (Sweden), a sole trader approved for F-skatt; organization and VAT numbers on every invoice.
- Your NDA signed and your document control followed; no professional indemnity insurance; laboratories in Sweden, Europe and the Asia-Pacific.