1,451 AI-Powered Medical Devices Cleared in the U.S. – Europe Lacks Even a Basic List

5 min read

By end of 2025, the U.S. Food and Drug Administration (FDA) had approved exactly 1,451 AI-based medical devices — 295 of them in 2025 alone. In Europe, no comparable public database exists. While American clinics deploy AI diagnostic tools on a daily basis, German hospitals rely on isolated solutions and wait for the EU AI Act to take hold — its requirements for high-risk systems won’t take effect until 2027. A status report.

295 authorizations in one year: what the FDA data actually reveals

Since 2017, the FDA has maintained a public database tracking every AI-based medical device that has received market authorization. The numbers tell a clear story: from 29 approvals in 2017 to 295 in 2025 — a new record that comfortably surpasses the previous high of 223 set just one year earlier.

76% of all FDA-authorized AI medical devices are in radiology — a field where pattern recognition across large imaging datasets performs especially well. GE HealthCare leads the rankings with 120 authorizations, followed by Siemens Healthineers with 89. The fact that two of the top three manufacturers have European roots makes the regulatory asymmetry between the United States and the European Union all the more striking.

Yet each authorization does not automatically translate into proven clinical utility. An analysis published in JAMA Network Open (April 2025), covering 903 FDA-authorized AI medical devices, arrived at a sobering conclusion: 24.1% had no clinical validation study whatsoever. Only 2.4% had been tested in a randomized controlled trial, and 8.1% in a prospective trial. That means roughly 90% of devices relied on retrospective data — or no clinical evidence at all.

This puts the figure of 1,451 FDA authorizations in perspective. As a measure of market momentum, it is impressive. As a quality benchmark, it falls short. For hospital procurement teams evaluating an AI-based system, FDA clearance alone is not a sufficient criterion. What matters far more is prospective studies conducted on comparable patient cohorts, along with transparent and reproducible clinical validation within their own care environment.

Europe’s blind spot: no database, no transparency

In Europe, no comparable public list exists. Neither the European Medicines Agency (EMA) nor Germany’s Federal Institute for Drugs and Medical Devices (BfArM) maintains a publicly accessible database of AI-based medical devices bearing the CE mark. The European database EUDAMED, intended to ensure transparency, has been in development for years. Some modules will become mandatory from May 2026, but a comprehensive overview of all AI-based products on the European market won’t be available for several more years.

Independent observers estimate there are at least 219 AI-based medical devices with CE marking (data from January 2025). The real figure is likely higher, but no one knows for sure. For hospital buyers facing the question of which AI system to evaluate, this presents a very real problem.

DACH Clinics Deploy AI—But Only as Isolated Solutions

Despite regulatory gaps at the European level, some DACH (Germany, Austria, Switzerland) clinics are already actively using AI-powered diagnostic tools. The results are promising—but fragmented.

Asklepios, one of Germany’s largest private hospital operators, has rolled out the Israeli AI system Aidoc across more than 25 facilities. The software analyzes CT scans in real time to detect strokes, intracranial hemorrhages, and vertebral fractures—funded through grants from the Hospital Future Act (KHZG). Its key advantage? Aidoc prioritizes imaging reports, ensuring critical cases are flagged for immediate radiologist review rather than lingering in the queue.

The largest prospective study to date on AI in diagnostics also comes from Germany. The Vara-PRAIM study, published in Nature Medicine, analyzed over 463,000 screening mammograms. The result: a 17.6% increase in breast cancer detection rates, alongside a reduced workload for radiologists. It remains the world’s largest prospective study of its kind.

“The question is no longer whether AI works in diagnostics. It’s now about how we can successfully bridge the gap from research to routine practice—regulatorily, organizationally, and economically.” — Prof. Alexander Berens, University of Tübingen (press release, 2025)

The catch? Every deployment is a standalone project. There’s no national standard for integrating AI into hospital information systems, no standardized interfaces, and no central registry showing which clinic uses which system. While the Federal Joint Committee’s (G-BA) Innovation Fund supports systematic AI integration in radiology through the xR.AI project, Germany is still far from widespread adoption.

The consequence? German clinics looking to introduce AI-based diagnostic tools must navigate every step themselves—from product selection to integration into their hospital information system (HIS), to training radiologists. Legal liability questions also remain unresolved at both national and European levels. If an AI system misses a diagnosis and the treating physician relied entirely on it, who bears responsibility? The Medical Devices Regulation (MDR) governs manufacturer liability, but the physician’s duty of care remains intact. In practice, this uncertainty leads many radiologists to avoid using AI findings as a decision-making basis, treating them at most as supplementary information.

The AI Act comes into force in 2027: What MedTech companies can expect

The EU’s landmark AI Act officially took effect in August 2024, but medical device manufacturers have been granted a transitional period: full compliance requirements for high-risk AI systems won’t apply until 2 August 2027. Any AI-based system classified as a medical device under the MDR or the In Vitro Diagnostic Medical Devices Regulation (IVDR) is automatically deemed high-risk.

What this means in practice: manufacturers must implement an AI-specific risk management system, meet stringent data quality and governance standards, produce technical documentation that goes beyond MDR requirements, and embed mechanisms for effective human oversight. The workload is substantial—especially for smaller players.

There is a glimmer of hope: in November 2025, the European Commission proposed an omnibus regulation aimed at streamlining procedures to reduce overlaps between the AI Act and MDR/IVDR. Instead of duplicating documentation and undergoing dual conformity assessments, an integrated process should be established. The regulation is expected to be adopted by summer 2026.

For Germany’s MedTech SMEs—93% of the sector employs fewer than 250 people—this is far from an academic concern. In recent years, the MDR alone has already driven some manufacturers to withdraw products from the European market, as certification costs outstripped revenues. If the AI Act adds another layer of bureaucracy, industry consolidation will only accelerate.

Yet the regulation also creates a qualitative edge: AI-powered medical devices that meet both MDR and AI Act requirements will enjoy a trust premium over those cleared only by the FDA. European manufacturers that clear both hurdles can leverage this dual certification as a global differentiator—provided they survive the financial and organisational challenges of the transition period leading up to 2027.

Accelerate or fall behind

The global AI healthcare market was valued at approximately $21.66 billion in 2025 and is projected to exceed $110 billion by 2030—a compound annual growth rate of 38.6% (MarketsandMarkets). PwC Strategy& estimates Europe’s potential at $203 billion.

Germany boasts strong AI-driven MedTech players, including Siemens Healthineers, Brainomix, Vara, and Merantix. Its clinical research is world-class, as the PRAIM study vividly demonstrates. What’s missing is the bridge between research and routine practice: a European AI database to ensure transparency, a harmonised approval process that doesn’t crush SMEs, and a hospital landscape that treats AI not as a subsidised pilot project but as a standard tool.

In eight years, the FDA has cleared 1,451 AI-based medical devices while refining its processes. Over the same period, Europe has drafted a regulation whose impact won’t be felt until 2027. Anyone who doesn’t see this lag as a strategic risk hasn’t grasped the market’s dynamics.

Frequently Asked Questions

Main image source: Pexels / MART PRODUCTION (px:7089623)