How it works

Assistive software for radiologists following colorectal liver metastases on contrast-enhanced CT. It does the finding and measuring; the radiologist decides.

What a follow-up read is made of.

Every number Scanvero reports sits on top of the layers beneath it. If a layer is missing or unconfirmed, nothing above it is shown.

  • Audit trailEvery confirmation and correction, recorded.
  • Radiologist confirmationEvery lesion link checked by a person. Phase confirmation is in development.
  • Lesion linksProposed only when the two scans align; otherwise "not available".
  • Liver and lesion outlinesFound and measured on every slice.
  • The CT studyStandard DICOM, read slice by slice.

Bottom layer: public research CT (CRLM-CT-1006). Upper layers are illustrations.

From two scans to a confirmed change.

The steps a follow-up read goes through, and who is responsible for each.

  1. Load the studies

    Standard DICOM CT from a patient's baseline and follow-up scans opens in a web-based reader. Measurement is withheld unless the contrast phase is confirmed: today from the scan's own record; radiologist confirmation is in development.
    Scanvero checks

  2. Find and measure

    Trained models outline the liver and each liver lesion, and measure lesion size and volume and liver volume, shown in axial, coronal and sagittal views.
    Scanvero proposes

  3. Propose links

    Lesions on the follow-up scan are matched to their counterparts on the baseline, but only when the alignment between the two scans passes its check. Otherwise the pair shows "not available".
    Scanvero proposes

  4. Confirm every link

    The radiologist reviews each proposed link side by side, and confirms, corrects or rejects it.
    Radiologist confirms

  5. Report the change

    For confirmed lesions only, Scanvero reports the change in size. Every decision is written to a tamper-evident audit trail.
    Scanvero records

Designed to hold back.

A follow-up measurement is only useful if it can be trusted. So Scanvero treats an unverified result as a failure, not a best guess.

  • Shows "not available" and the reason, never a hidden warning beside a confident number.
  • Reports change only for lesion links a radiologist has confirmed.
  • Keeps every result traceable to the source images it came from.
  • Records confirmations and corrections in a tamper-evident audit trail.

Intended use, today.

Scanvero software is for research use only. It is not yet UKCA or CE marked and must not be used for clinical decisions.

It is being developed as software intended to become a medical device, with risk management (ISO 14971) and software lifecycle (IEC 62304) practice in mind. A regulatory route will be defined with clinical partners before any clinical use.

Scope is colorectal liver metastases on portal venous contrast CT. Scans outside that scope are meant to be refused, not measured.

Work with us

We are looking for hospitals and research groups with baseline and follow-up CT of colorectal liver metastases, radiologists for a reader study, and academic and NHS partners for funded validation.

info@scanvero.co.uk
Email usAbout Scanvero