Radiology accounts for 76% of all FDA-cleared AI/ML-enabled medical devices as of the end of 2025.
Key Details
- 1A longitudinal analysis covered 1,430 FDA-authorized AI/ML medical devices between 1995 and late 2025.
- 2Annual authorization of such devices jumped from 1.8 (1995-2014) to 264 per year (2023-2025), with 331 approvals in 2025 alone.
- 3Radiology devices represented 76.5% (1,094 of 1,430) of all authorized devices; cardiovascular and neurology followed.
- 4Most devices (96%) went through the streamlined 510(k) approval pathway; De Novo and PMA routes were rare.
- 5740 companies contributed devices, but only 13 made up 17.3% of authorizations, indicating market concentration.
- 6Several clinical specialties (e.g., pathology, OB/GYN, psychiatry) had very few or no authorized AI devices.
Why It Matters
This analysis underscores radiology's central role in AI medical device innovation and regulatory clearance, shaping the present and future landscape of imaging AI in medicine. The findings highlight both rapid progress and areas where broader adoption is needed, informing researchers, vendors, and policymakers.
Source
AuntMinnie
Related News
•Radiology Business
Automation Bias: How AI Can Compromise Radiologist Accuracy
AI decision support can induce automation bias, leading radiologists to accept incorrect interpretations and potentially reduce their diagnostic accuracy.
•AuntMinnie
Study Reveals Major Impact of Incorrect AI Suggestions in Mammography Reads
Incorrect AI suggestions, especially false-negatives, significantly reduce reader sensitivity and alter visual search in mammography interpretation.
•HealthExec
Mayo Clinic Sued Over Alleged AI Improprieties and Whistleblower Retaliation
A former Mayo Clinic research director sues the institution, alleging retaliation after raising concerns about improper AI use affecting patient safety and data integrity.