Most AI diagnostics companies win funding by finding more disease; Eyedentity raised its first round by promising to send fewer people to the hospital.

The Stockholm company said on Aug 21, 2026 that it had closed an oversubscribed first investment round of EUR 1.3 million, roughly $1.5 million or SEK 15 million. Norrsken Launcher led, with Karolinska Institutet Innovation and a group of angel investors participating. No valuation was disclosed and no earlier institutional funding has been reported. A global deal feed listed the round at $2 million; the figure in press coverage is EUR 1.3 million.

What the company does

Eyedentity was founded in 2025 by Gustav Stalhammar, an ocular oncologist at St. Erik Eye Hospital and professor at Karolinska Institutet, and Mats Holmstrom, a former machine learning specialist at RaySearch Laboratories. Stalhammar is chief scientific officer; Holmstrom is chief executive.

The product reads fundus images — the photographs of the back of the eye taken during an ordinary eye examination — and flags signs of serious disease. Its first target is uveal melanoma, a rare eye cancer whose outcomes depend heavily on how early it is caught. The commercial premise is that the imaging infrastructure already exists in high street optician chains, so the software rides on scans that are being taken anyway rather than requiring new hardware or a new patient visit.

The numbers behind the round

The metric the company leads with is not detection rate but restraint: Eyedentity says its system can cut false-positive referrals by 90 percent. That framing is aimed squarely at the bottleneck in Swedish specialist ophthalmology, where suspicious findings from opticians generate referral volume that clinics struggle to absorb.

It is running a live pilot with Synologen, a Swedish optician network. Holmstrom has described the integration target as zero clicks for the optician — the model has to disappear into existing workflow to be adopted at all. Anna Fredrixon, a partner at Norrsken Launcher, framed the thesis in the announcement as technology that can save lives without sending more patients into an already overburdened healthcare system. The money is earmarked for regulatory approval, further diagnostic capabilities beyond the initial indication, hiring, and preparing for commercial deployment.

Why the timing matters

European medical AI is in a phase where regulatory clearance, not model performance, is the gating factor, and a $1.5 million first round is a runway to a submission rather than to a market. That is the honest read of this deal size: it buys the company its regulatory dossier and a pilot-to-contract conversion at Synologen.

The Karolinska connection matters commercially as well as scientifically. A clinician co-founder with a specialist practice and an institutional investor attached to the same academic system shortens the distance between a research claim and the clinical validation that Swedish procurement will ask for. Whether the 90 percent figure holds outside a pilot is the question the next round will be priced on.

Source

EU-Startups — "Stockholm-based Eyedentity secures EUR 1.3 million to detect deadly eye cancer earlier with AI"