Trust and safety
Software that measures cancer has to earn trust before it earns use. This page sets out how Scanvero is built to be safe, how data is handled, and exactly where we stand on regulation.
Safety: fail closed
- A result that cannot be verified is shown as "not available" with its reason, never as a number with a warning beside it.
- Change is reported only for lesion links a radiologist has confirmed.
- Scans outside the intended scope are meant to be refused, not measured.
Data
- Partner scans are shared only under a written agreement with the partner's approvals.
- Hospitals remove patient names, NHS numbers and other identifying details from scans before sharing them, and we use them only for the study we agreed.
- This website sets no cookies and runs no analytics. Our host keeps standard server logs, as our privacy notice explains.
Security and traceability
- Every measurement points back to the source images it came from.
- Confirmations and corrections are written to a tamper-evident audit trail.
Regulatory status
- Research software for research use only. Not yet UKCA or CE marked, and not for clinical use.
- We already work to documented engineering practice: pass marks fixed before every test, sealed held-out data, version-controlled code and archived models, and risk management with ISO 14971 in mind.
- A formal quality management system (ISO 13485) and the UKCA regulatory route will be in place before any clinical use.
What we will not do.
- Show a number we cannot trace to the images it came from.
- Use a partner's data beyond the study it was shared for.
- Claim clinical performance we have not measured, or quote a result without the data it was measured on.
- Choose a pass mark after seeing the result.
Questions we are asked.
If yours is not here, email info@scanvero.co.uk.
Is Scanvero a medical device?
Not yet. It is research software for research use only. It is being developed as software intended to become a medical device, and it will need UKCA marking before any clinical use.
Can we try it?
We are not offering clinical use. We are looking for research partners to evaluate it on their own follow-up scans under a written agreement. See Partners.
What data would you need from us?
Baseline and follow-up CT of colorectal liver metastases, with patient names, NHS numbers and other identifying details removed, and the reference reads needed for the agreed study. Nothing is shared until the agreement is in place.
Does it replace the radiologist?
No. It proposes outlines and lesion links. A radiologist confirms each link before any change is reported, and every decision is recorded.
How good is it?
Component results on public research data are on the Evidence page, with the dataset for every figure and a list of what has not yet been shown.
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.
Please do not email patient images or patient information.