Oxipit
Oxipit is a CE-marked radiology AI suite for chest X-ray, chest CT and musculoskeletal imaging. Its ChestLink module reports healthy chest X-rays autonomously, clearing up to 40% of the worklist without radiologist involvement.
What is Oxipit?
Oxipit is a radiology AI vendor based in Vilnius, Lithuania, whose software reads medical images and then writes or checks the report that goes with them. The company grew out of a 2017 AI hackathon, where a team of clinicians and data scientists set out to attack the global shortage of radiologists. Its first product, ChestEye, was CE-certified in 2019 and produced preliminary reports across 75 chest X-ray findings.
The product that made Oxipit's name arrived in April 2022. ChestLink is presented as the world's first CE Class IIb-certified autonomous AI application for healthy chest X-ray reporting: it identifies normal studies with a claimed 99.9% precision and issues the final report itself, removing up to 40% of cases from the worklist without a radiologist ever opening them. That autonomy is CE marked and available in Europe only — the company states plainly that autonomous use is not available in the United States.
The portfolio now spans three suites. CXR Suite combines ChestLink, ChestEye and a Quality module that re-reads finalised reports as an automated safety net. CT Suite, powered by CT Eye, covers pulmonary embolism, lung lesions and consolidation on adult chest CT, with nine further findings announced. MSK Suite, powered by MSK Eye, reads musculoskeletal X-rays for fractures, arthritis, osteoporosis, prostheses and more.
Everything is built to run inside the department's existing systems. Oxipit integrates with PACS and RIS over DICOM and HL7, deliberately operating under the surface with no extra dashboard or screen, and reports come out in the institution's own language across twelve European languages. Deployment is usually on the customer's premises, with cloud available on request. The company quotes an average area under the curve of 93% and productivity gains of at least 30%.
Oxipit is a regulated medical device, restricted to healthcare professionals and not sold to the public. It has been part of the Swedish imaging group Sectra since an acquisition agreed in March 2026, and is also distributed through the Blackford, deepc, CARPL and Newton's Tree platforms.
What it does
- Detect 75 different findings on chest X-rays
- Report healthy chest X-rays autonomously, with no radiologist in the loop
- Generate preliminary reports in the institution's own language
- Triage and prioritise the reading worklist
- Run a real-time quality check on reports after they are signed off
- Flag pulmonary embolism, lung lesions and consolidation on chest CT
- Detect fractures, arthritis and bone lesions on musculoskeletal X-rays
When to use Oxipit / When not to
A quick filter to help you decide if Oxipit is the right fit.
When to use Oxipit
- High-volume radiology departments handling anywhere between 5,000 and 500,000+ chest X-rays a year
- Hospitals and imaging networks struggling with radiologist shortages and reporting backlogs
- Radiology teams already running a PACS or RIS that speaks DICOM and HL7
- Multi-site diagnostic providers standardising chest imaging across a whole network
- Services that want an automated quality-assurance layer re-reading finalised reports
When not to use Oxipit
- Individuals and general consumers: the software is not sold to the public, only to healthcare organisations
- US sites hoping to run autonomous reporting, which Oxipit explicitly excludes outside Europe
- Departments needing modalities beyond chest X-ray, adult chest CT and musculoskeletal X-ray
- Facilities without a PACS or RIS, or without the ability to integrate over DICOM and HL7
- Buyers who need public pricing or a self-service sign-up before talking to a salesperson
How to use Oxipit
A typical end-to-end flow, from setup to results.
- Request a demo through the contact form on the Oxipit site — there is no online sign-up
- Prepare a set of your own chest X-rays so they can be analysed in real time during that demo
- Ask for extended platform access if you want a longer clinical evaluation first
- Agree a quote with the sales team, since no price is published anywhere
- Choose a deployment model: on-premise by default, cloud where client and legislative requirements allow
- Connect the software to your PACS and RIS over DICOM and HL7
- Pick the workflow depth you need, from real-time notifications up to full autonomy on normal chest X-rays
- Have the report templates customised to your institution's reporting conventions
- Confirm the reporting language, or request a new one — Oxipit quotes roughly four weeks to add one
- Go live, which Oxipit says takes under a week with minimal IT setup
Pros & Cons
Pros
- The only suite claiming certified autonomous reporting of healthy chest X-rays, at CE Class IIb
- Broad coverage: 75 findings, described as roughly 90% of what a department sees day to day
- Requires no workflow change — it runs inside the existing PACS or RIS with no extra screen
- On-premise deployment means patient images need never leave the hospital
- Certifications are published as downloadable PDFs and can actually be checked, not merely claimed
- Reports are produced in the institution's own language and tailored to local conventions
- Backed since 2026 by Sectra, an established enterprise imaging group
Cons
- No published pricing whatsoever — every enquiry has to go through a quote
- The site publishes no terms and conditions, so the contractual framework cannot be reviewed online
- Autonomous operation is unavailable in the United States
- The 32 European markets and 12 supported languages are advertised but never actually listed
- The privacy notice covers the website only and says nothing about clinical image processing
- That same notice still relies on the EU-US Privacy Shield, a framework invalid since 2020
- CT Eye is limited to adult chest studies, and several announced findings have not yet shipped
Pricing & Plans
Oxipit publishes no pricing. There is no permanent free plan and no free trial: the FAQ answers the question "What is the price?" four separate times by directing the reader to a quote request. A demonstration is offered free of charge on request, and extended platform access may be granted for clinical evaluation, but no amount, currency or billing unit appears anywhere on the site. Pricing is handled entirely through a sales conversation.
Data, GDPR & hosting
A consolidated view of how Oxipit handles your data.
GDPR overview
Oxipit claims GDPR compliance explicitly, though in its FAQ rather than on a dedicated compliance page: the software works with anonymised data "in order to fully comply with GDPR regulations", and images need not leave the client's infrastructure. The website privacy notice rests on recognisable GDPR foundations — a named controller, stated legal bases (legitimate interest for correspondence, consent for cookies), the rights of access, rectification, erasure, restriction and objection, and a right to complain to any EEA supervisory authority. No DPO is named; info@oxipit.ai is the single contact point. Two gaps deserve mention: the notice still cites the EU-US Privacy Shield to justify transfers to Formspree and Google, a framework invalidated in 2020, and it covers the website alone, never the processing of clinical images.
Who owns the data?
For the website, Oxipit, UAB (company code 304617386, Saulėtekio 15, Vilnius) is the named data controller and is responsible for personal data submitted through the contact form or by email. That published notice covers website traffic only: it says nothing about ownership of the clinical images the products process. In practice the software is normally deployed on the customer's own premises, with processing and storage staying on site, and Oxipit states that images need not leave the client's IT infrastructure at all. Patient data is described as anonymised. No contractual terms governing image ownership are published anywhere on the site.
Reuse rights
No terms and conditions are published on the site at all, so the permissions a contract would normally set out — what a customer may reuse, and on what conditions — are simply not available online. What the privacy notice does describe is Oxipit's own use of the personal data it collects: contact details and message content gathered through the form or via info@oxipit.ai are used to evaluate potential partners, manage correspondence and answer enquiries, on a legitimate-interest basis. Cookies are set only with consent, Google Analytics measures browsing behaviour, and the notice reserves the right to pass personal data to affiliates in a merger, acquisition or similar transaction.
Data retention & training
Hosting summary
Hosting is largely the customer's affair. Oxipit's products are usually deployed on the client's own premises, with processing and storage staying on site, and the company states that images need not leave the client's IT infrastructure at all. Cloud deployment is offered as an alternative where client requirements and local legislation allow. Data is described as anonymised before processing. No hosting country or region is declared for the clinical products, so jurisdiction follows wherever the customer installs the software rather than a location Oxipit picks. The picture differs for the website, which is all the published privacy notice actually covers. Contact-form submissions are handled by Formspree, Inc. of San Antonio, Texas, and browsing statistics by Google LLC of Mountain View, California — both United States companies, with the notice justifying those transfers by reference to the EU-US Privacy Shield, a framework invalidated in 2020. Network lookups place the oxipit.ai web server itself with Hetzner in Nuremberg, Germany, though the company does not state this.
Where Oxipit works
Country-level availability.
Not available in
Things to keep in mind
Risks and trade-offs to weigh before adopting Oxipit.
- Autonomous reporting means a share of studies is never seen by a human. The safety case rests on the 99.9% figure holding in your patient population, not the vendor's.
- Automation bias is a real clinical risk: radiologists who come to trust an AI safety net can read less carefully, and the effect is hard to notice from inside.
- Skill erosion matters for training. Junior radiologists who never see the normal studies ChestLink absorbs lose much of their calibration for what normal looks like.
- The legal entity was renamed at the Lithuanian register: company code 304617386 is now filed as Sectra Lithuania, UAB while the site still says Oxipit, UAB. Contracts should name the right one.
- No terms and conditions are published, so liability, service levels and the division of clinical responsibility cannot be assessed before entering a sales process.
- Certificates carry expiry dates — the CE certificate to 22 February 2027, the ISO 13485 certificate to 21 November 2027 — and should be rechecked rather than assumed current.
- Regulatory approval varies by market and not all features ship everywhere; assuming a capability seen on the website is available locally is a mistake.
Setup & Integrations
Technical difficulty
Moderate, and mostly an integration exercise rather than a software one. The product connects to an existing PACS or RIS over DICOM and HL7, and a DICOM conformance statement is published for that interface. Oxipit says there is no extra dashboard to deploy, no change to the department's workflow, and go-live in under a week with minimal IT setup. Installation is on-premise by default, cloud on request, and report templates are customised to local conventions. A lighter route exists for sites already using Blackford, deepc, CARPL or Newton's Tree.
Deployment
Integrations
Behind Oxipit
Fundraising
Social
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What does Oxipit actually do?
What makes ChestLink different from other radiology AI?
How many findings does it cover?
How much does Oxipit cost?
How does it fit into our existing systems?
Is it cloud-based or installed on site?
What happens to patient data under GDPR?
Can we use it in the United States?
Which languages are the reports written in?
Who is behind Oxipit?
Should you pick Oxipit?
Oxipit sits at the serious end of medical imaging AI. This is not a tool anyone signs up for on a Tuesday afternoon: it is a regulated medical device, sold to healthcare organisations through a quote, deployed inside a hospital's PACS, and covered by certificates you can download and read — an ISO 13485 quality system and a CE certificate under Regulation (EU) 2017/745, both issued by TÜV Rheinland.
What genuinely distinguishes it is autonomy. Plenty of vendors offer detection and triage; ChestLink goes further and issues the final report on normal chest X-rays with no radiologist in the loop, which Oxipit says clears up to 40% of the worklist. For a department drowning in volume that is a different proposition from a second opinion. The caveat is equally clear, and comes from Oxipit itself: that autonomy is a European capability, explicitly not available for autonomous use in the United States.
The gaps are mostly about disclosure rather than the product. No prices are published, no terms and conditions exist on the site at all, and the headline numbers — 99.9% precision, 93% AUC, 30% productivity gains, 32 European markets, 12 languages — are vendor claims, never itemised or independently sourced. The privacy notice covers the website rather than the clinical images, and still cites the EU-US Privacy Shield, struck down in 2020.
Since March 2026 Oxipit has been part of Sectra, which gives the technology a far larger distribution footprint and a buyer rather more institutional reassurance. If you run a high-volume chest imaging service in Europe and can support a PACS integration, Oxipit is one of the few credible routes to genuinely autonomous reporting. Expect a procurement conversation, not a trial link.
- Choosing a selection results in a full page refresh.
- Opens in a new window.