An analysis found that out of 1,357 AI-enabled medical devices cleared by the FDA, only three were actually tested for their effect on patient health outcomes.
Key Details
- 11,357 FDA-cleared AI medical devices were analyzed.
- 2Only 34 devices had associated registered clinical trials.
- 3Just 3 devices were tested for patient-centered outcomes like survival or quality of life.
- 4Most studies excluded groups like pregnant women, older adults, and non-English speakers.
- 5Current FDA clearance requires 'substantial equivalence,' not proven patient benefit.
- 6Authors propose new frameworks for requiring outcome-based evidence across diverse populations.
Why It Matters
Most AI devices, including those used in imaging and diagnostic support, are authorized without proof that they improve real-world patient outcomes. This policy gap is significant for radiology, as clinical validation is crucial for responsible AI deployment and patient safety.
Source
EurekAlert
Related News
•EurekAlert
AI Framework Accelerates Aortic Aneurysm Risk Prediction from Imaging
Researchers developed BioPINN-LM, combining physics-informed neural networks and multimodal large language models to deliver fast, interpretable risk assessments for ascending thoracic aortic aneurysms.
•EurekAlert
Study Finds Patient Voices Missing in Generative AI Design for Oncology
A Flinders University-led review found patients and carers are rarely involved in shaping generative AI tools used in oncology.
•EurekAlert
Scripps Researchers Develop AI Foundation Model for ECG-Based Heart Disease Prediction
A new AI foundation model, ECG-CLIP, improves detection and prediction of multiple heart diseases using large-scale ECG and clinician note data.