IN DEVELOPMENT
Track™
Reads every radiology report and pulls out the findings and the follow-ups they recommend. Then it watches whether each follow-up actually happens, and flags the ones that don't.
Clinical memory for radiology.
indelva is a clinical memory platform. It follows every important finding across reports, scans and letters, so nobody has to piece the history back together.
Now seeking clinical partners to trial our technology.
The information is already there. Finding it is the problem.
In healthcare, the problem isn't a lack of information. It's that the information clinicians need is buried, fragmented and hard to find.
0 billion
imaging studies performed globally each year
0
clinician-years spent every year searching for context that already exists
0%
shortfall of UK radiologists today, rising to 40% within three years
Healthcare should not have to rediscover the same finding every time someone opens the record.
A summary is not the same as memory.
Most AI tools in healthcare summarise. A summary is new text, generated fresh each time, with nothing tying it back to what the record actually says. indelva is built differently.
A summary
- —Rebuilt from scratch each time
- —Ask twice, get two different answers
- —No link back to the source
- —Omissions are invisible
A memory
- —Persists between clinicians
- —Every item linked to its exact source words
- —Represents what the record says, never infers
- —Auditable, every time
In clinical work, an answer you cannot trace is an answer you cannot trust.
A new unit of the medical record
One lived example: a lung nodule tracked across four clinical documents over a year.
TODAY
time-based record
May 2023
6mm solid nodule, right upper lobe
source: radiology report
Nov 2023
Stable appearances
source: clinic letter
Mar 2024
6.8mm
source: third-party AI measurement
May 2024
Nodule not mentioned
Was it resolved, or missed?
source: radiology report
re-organised around the finding
INDELVA
finding-based record
Lung nodule, right upper lobe
6.0 mm → 6.8 mm growing
May 2023radiology report
Nov 2023clinic letter
Mar 2024AI measurement
May 2024radiology report
May 2023 – May 2024 · 4 sources · 1 gap
One finding, its whole history in one place.
Four documents. One finding. When a finding simply stops being mentioned, nothing tells you whether it resolved or was missed.
We move from time-based entries to finding-based memory.
Every finding. One memory.
Everything indelva knows about a finding flows into one place.
THE CORE
Insight™
Brings together everything known about a finding, across reports, letters and images, into one continuous history. It governs what changes, what stays uncertain, and what needs review.
ON THE ROADMAP
Vision™
An image-level layer, bringing imaging AI outputs into the same record. On the roadmap, planned for future development with partners, including imaging AI from Raida.
Who we're talking to.
is developing its first product, supported by Innovate UK's ICURe programme, following market discovery across nine countries and over 100 conversations.
We're speaking with hospital innovation teams, lung cancer screening providers, and PACS and reporting partners.
Speak to us.
A 15-minute conversation is enough to start. We'd like to understand how this problem shows up in your work, and what would make it worth solving.