Diagnostics & imaging AI · CT radiology AI
Avicenna.AI
Avicenna.AI builds FDA-cleared CT radiology AI (CINA portfolio) for emergency and incidental findings, with orchestration meant to deliver results inside existing PACS/RIS workflows. It is not a POCUS-only or ambient scribe product.
Strong fit
- Hospitals and remote radiology groups that need FDA-cleared AI to flag urgent CT findings such as stroke, pulmonary embolism, and spine injuries
- Teams that can send results into their current PACS and RIS, rather than adding another viewer for staff to watch
- Buyers comparing focused CT AI products, not AI note-writing tools
Weak fit
- Sites without enough emergency CT scans to justify triage AI
- Buyers who need bedside ultrasound AI, like AISAP, or AI scribes
- Organizations that will not assign staff to review alerts and handle false alarms
Bottom line
Avicenna.AI is worth considering for imaging programs that specifically want AI to flag urgent and incidental findings on CT scans. It suits radiology teams ready to tune alerts so readers are not flooded. Large U.S. health systems that want broader radiology AI from a bigger vendor should look at Aidoc.
Score breakdown
Weights: Outcomes 35% · Product 30% · Implementation 20% · Pricing clarity 15%.
Avicenna.AI builds FDA-cleared radiology AI for CT scans. Its CINA modules flag emergency and incidental findings. The company has multiple FDA clearances and about 300 installed sites.
In 2025 it added AVI, which delivers results inside the PACS and RIS systems radiologists work in.
The results depend on whether alerts change what happens to patients without flooding readers with noise. Its product is competitive in emergency CT AI, but it is narrower and smaller than Aidoc, which many large U.S. health systems consider.
Implementing it is a radiology IT project, and prices are set for each customer.
Other products fit different jobs. Aidoc suits health systems that want broader radiology AI, AISAP suits bedside ultrasound, and Studycast suits cloud imaging workflow when triage alerts are not the main need.
Among the imaging AI tools we review, Avicenna.AI scores just below AISAP and above Ferrum. Its Conditional rating reflects its smaller U.S. scale, narrower coverage, and unclear pricing more than any weakness in the clinical idea.
Competitor landscape
| Vendor | Overall | Ease of implementation |
|---|---|---|
| Aidoc | 7.9 | 7.5 |
| ScreenPoint Medical | 7.9 | 7.6 |
| Harrison.ai | 7.7 | 7.5 |
| Qure.ai | 7.4 | 7.1 |
| Riverain Technologies | 7.3 | 6.9 |
| Subtle Medical | 7.2 | 6.8 |
| DeepTek | 7.1 | 6.8 |
| AZmed | 6.9 | 6.7 |
| Enlitic | 6.9 | 6.5 |
| Avicenna.AI | 6.9 | 6.6 |
| Ferrum Health | 6.8 | 6.5 |
| Koios Medical | 6.7 | 6.3 |
Pricing
| Item | Detail |
|---|---|
| Model | Software licenses for the CT AI modules and the layer that routes results, priced by site and volume. |
| What usually drives cost | Which cleared modules you need, the number of sites, scan volume, and how results are routed. |
| What to ask before you buy | The cost per site for the CINA modules you will turn on, how much AVI connection work is involved, and how false alarms will be handled. |
| Published pricing | Avicenna.AI does not publish prices. Expect a quote for your radiology AI program. |
Match FDA-cleared indications to the CT exams you run before enterprise rollout.
Prerequisites for purchase
| Need | Why it matters |
|---|---|
| What you need to get Avicenna.AI to function | |
| A clinical owner in radiology or cardiology, plus imaging IT | These tools fail without both. |
| A written plan for connecting to PACS, RIS, and the EMR | Results that land where no one looks create a safety risk. |
| Agreement on which kinds of scans are covered | Trying to cover every kind of scan at once delays results. |
| A plan for who reviews alerts and reports | Lists no one owns get ignored. |
| Agreed times for changes that could interrupt clinical work | New screens that appear without warning during busy hours destroy trust. |
| What will maximize your value | |
| Measure turnaround and how often reports are reopened or edited | Accuracy in a demo does not prove it works day to day. |
| Start with one common type of scan | A small real win is worth more than a broad launch with nothing behind it. |
| Review false positives weekly with radiologists | Tools that raise too many false alarms get turned off. |
| Keep radiologists in native viewers when possible | Extra reading lists drive radiologists away. |
| Check which of your scan types the FDA clearances cover | Paying for clearances that do not match your scans wastes money. |
| Deal-breakers | |
| No imaging IT or clinical owner. | |
| You will not do any connection work. | |
| You expect imaging AI to write clinical notes. | |
| Nobody will review alerts or structured reports. | |
| Your legal team blocks the cloud or vendor handling of patient data that you need. | |
Value creation time frame
| # | Stage | Typical range |
|---|---|---|
| 1 | Scope | 2-4 weeks. Agree on scan types, connections, and how success is measured. |
| 2 | Integrate | 4-12 weeks. Connect to PACS, RIS, and the EMR, and test on real cases. |
| 3 | Pilot | 4-8 weeks. Run at one site or for one kind of scan. |
| 4 | Expand | Ongoing. Add sites and adjust alerts. |
Leadership
Methodology
| Weight | Factor | What it measures |
|---|---|---|
| 35% | Customer outcomes | Whether buyers get measurable operational or clinical-workflow results after go-live |
| 30% | Product | Capability depth, reliability, and fit for the job the category buys |
| 20% | Implementation | How hard it is to stand up, integrate, train, and stabilize |
| 15% | Pricing clarity | Whether a buyer can model total cost without a mystery quote |
| Label | Meaning |
|---|---|
| Highly recommend | Strong outcomes and product with manageable caveats |
| Recommend | Solid fit for the right buyer; know the tradeoffs |
| Conditional | Only with a specific use case or heavy caveats |
| Not recommended | Avoid for most buyers in this category |
Read our full methodology for how we weight scores and assign recommend labels.