Upload a CT
Drop in an abdominal CT series. The model needs the CT alone: no report, no manual markup.
Research preview · Built in Sydney
We're building toward a viewer for radiologists: upload an abdominal CT and see the pancreatic tumour highlighted with a soft, shape-faithful contour, and an honest band where the edge is uncertain. We're in research preview, and we publish our evidence as we go.
Research preview · not for clinical use
Illustration with a synthetic slice. See it on real CT in the live demo →
The product we're building
The workflow we're designing toward. Each step maps to a piece of research that is under way; none of it is a finished clinical product.
Drop in an abdominal CT series. The model needs the CT alone: no report, no manual markup.
Our model traces the lesion, including where it extends past the pancreas.
See the most likely boundary, plus a graded band where it's uncertain. It supports the radiologist's delineation. It is not a diagnosis, and the radiologist stays in charge.
Pricing
No seats, no licence fees, no minimums. A radiology team pays for the scans it contours, and nothing when it doesn't.
One CT in, one tumour contour and uncertainty band out.
Why the edge
Pancreatic cancer is infiltrative: it wraps vessels and often extends past the gland. A useful outline has to show where it really ends.
Most pipelines approximate the tumour's shape, which works for finding it. Then they cut away anything outside the pancreas, which removes real tumour at exactly the edge that matters.
The outline ignores the tumour's real shape, and tissue outside the organ is deleted.
A contour that follows the tumour's real edge, with anatomy guiding the model instead of amputating tumour.
Evidence so far
Every number below comes from our own experiments, labelled with the study and cohort it came from. We don't have clinical accuracy figures yet, and we won't invent them.
development cases with tumour had reference tumour outside the supplied pancreas mask: 74% of reference volume. A hard organ cut would delete it.
Internal study · 16 development cases
of expert-labelled tumour can fall outside the region a standard pipeline lets the model search (26 of 143 cases).
Internal study · 143 training cases
change in overlap from simply widening the search. That's far short of the gain we set in advance, so the fix has to be smarter.
Internal study · 143 cases
Roadmap
From one model to the tool radiologists open for every tumour they read.
The model that traces a tumour's real edge, not a sphere.
Upload a CT, get the tumour highlighted with its uncertainty band.
First partner teams on per-scan pricing, shaping the viewer with us.
DICOM and PACS integration, so contours open where radiologists already read.
Size, volume and vessel contact from the contour, ready for the report.
Compare contours across scans to see growth or response to treatment.
From pancreas to liver, kidney and beyond.
Validated and cleared for clinical use: the contour every radiologist trusts.
Team
One team with one goal: show radiologists where a tumour really ends. Engineers and researchers building in Sydney.
Meet the teamRadiologists wanted
If you read abdominal CT, twenty minutes of your opinion on how a contour and its confidence band should be shown would shape what we build next.