A new study has found that 99.8% of AI-based medical devices cleared by the FDA have never been tested to determine whether they improve patient outcomes.
Published in PLOS Digital Health, the study highlights major gaps in the FDA’s current approach to evaluating these technologies.
Using data from the FDA device database and the ACR Data Science Institute catalogue, the researchers conducted a systematic analysis of all FDA-cleared AI medical devices through December 2025.
Out of 1,357 cleared devices, only 34 (2.5%) were linked to registered prospective trials, and only three of these evaluated patient-centered outcomes such as death rates, hospital readmissions, and quality of life.
Based on these findings, the scientists propose a roadmap for strengthening the clinical validation of AI medical devices.
This includes mandatory prospective studies conducted in real-world clinical settings, as well as post-clearance requirements for multicenter trials with patient-centered endpoints and prespecified subgroup analyses.
【MORE】