Researchers unveiled a new taxonomy and online tool for classifying 1,000+ FDA-cleared AI/ML-enabled medical devices, with radiology as the dominant specialty.
Key Details
- 1A database and online tool summarize 1,016 FDA-cleared AI/ML devices through December 20, 2024.
- 2736 unique devices were identified after consolidating multiple authorizations per device.
- 3Most devices (88.2%) are built for radiology; neurology (2.9%) and hematology (1.9%) follow.
- 4Clinical function analysis: 84.1% assist in assessment, not intervention; 85.6% of AI functions are analysis, mostly for quantification/feature localization (65%), and triage (12.9%).
- 5Only three devices use EHR/tabular data, showing clear dominance of imaging data inputs.
- 6Key application types, definitions, and usage patterns are detailed in the report and a public website.
Why It Matters
This resource offers unprecedented clarity about the distribution, function, and data types of AI radiology devices, supporting better validation strategies and more informed clinical adoption. It empowers radiology professionals, vendors, and regulators to understand which AI tools exist, what they actually do, and where gaps remain.
Source
AuntMinnie
Related News
•AuntMinnie
Paul Chang Discusses Foundation Models and Agentic AI at RSNA 2025
Dr. Paul Chang shares his insights on the role of foundation models and agentic AI in radiology at RSNA 2025.
•AuntMinnie
Toronto Study: LLMs Must Cite Sources for Radiology Decision Support
University of Toronto researchers found that large language models (LLMs) such as DeepSeek V3 and GPT-4o offer promising support for radiology decision-making in pancreatic cancer when their recommendations cite guideline sources.
•AuntMinnie
AI Model Using Mammograms Enhances Five-Year Breast Cancer Risk Assessment
A new image-only AI model more accurately predicts five-year breast cancer risk than breast density alone, according to multinational research presented at RSNA 2025.