Yes. MedTech Dive reported on September 18, 2026, that Cognita Imaging received a $1.29 million Food and Drug Administration research contract to test whether a “jury” of large language models can evaluate AI-generated radiology reports and flag cases that need radiologist oversight. The contract began June 22 and runs for 18 months; Cognita is working with the FDA Center for Devices and Radiological Health’s regulatory science team and plans to apply the framework to about 1 million patient exams.
The project aims to measure clinically meaningful hallucinations and omissions in generative radiology tools and to give the FDA reusable guidance, software code, and discrepancy analyses for regulating generative AI devices.
Source: Cognita Imaging wins FDA contract to test LLMs in evaluating radiology AI — MedTech Dive