Less than 30% of FDA-cleared AI devices shared key safety and adverse event data before approval, raising regulatory concerns.
Key Details
- 1A new analysis reviewed FDA approval and recall databases for all AI/ML devices cleared from 1995 to July 2023.
- 2Of 691 devices analyzed, 531 targeted radiology applications.
- 3Fewer than 30% of devices provided data on safety, efficacy, or adverse events prior to approval.
- 4No predefined efficacy or safety reporting standards currently exist for AI/ML devices in the U.S., unlike with pharmaceuticals.
- 5The findings highlight a regulatory gap and call for stricter testing and dedicated approval pathways for AI/ML-based medical devices.
Why It Matters
This research underscores a major gap in regulatory standards and transparency for AI devices in radiology, potentially impacting patient safety and trust. Addressing these concerns could lead to more robust evidence and safer adoption of AI in clinical imaging practice.
Source
Radiology Business
Related News
•Radiology Business
AI Guidance Cuts Novice Ultrasound Exam Time by 34%
AI guidance significantly reduces exam times and enhances diagnostic quality for novice ultrasound operators performing shoulder exams.
•Radiology Business
UCLA Appoints Inaugural Associate Dean for Health AI Strategy
UCLA has appointed Katherine P. Andriole as its first associate dean for Health AI Strategy and Innovation, with an initial focus on radiology.
•AuntMinnie
AI Models Reveal Racial Disparities in Breast Cancer Patterns
Machine learning models reveal significant racial disparities and key predictors in breast cancer incidence across diverse groups.