FDA clears RadNet's foundation model-based chest X-ray AI tool
RadNet subsidiary DeepHealth has received 510(k) clearance from the U.S. Food and Drug Administration for an AI-powered chest X-ray solution designed to detect and localize suspected thoracic abnormalities on radiographs.
The newly cleared tool, Chest XRay, is built on a foundation model trained across large volumes of chest images. DeepHealth says the approach is intended not only to support current computer-aided detection capabilities, but also to provide a foundation for developing additional findings and workflow tools. The same underlying model is already used in this way in Europe through the CE-marked version of Chest XRay. There, the company says the model analyzes more than 2.8 million examinations annually.
“Foundation models are important building blocks,” Sham Sokka, PhD, RadNet's chief operating and technology officer, digital health, said in an announcement Sept. 16. “Chest XRay demonstrates how a broadly trained foundation model can be translated into a precise, FDA-cleared clinical capability today, while establishing a platform for additional intelligent interactive experiences over time. This clearance marks an important step in DeepHealth’s vision of bringing intelligence into every stage of the imaging experience.”
The software is designed to identify and localize four categories of thoracic findings on chest radiographs: nodules, consolidation, mediastinal and hilar abnormalities, and pleural space abnormalities, including pneumothorax. It is intended to support radiologists across routine chest X-ray interpretation as well as more urgent exams by flagging suspicious findings for further review.
The new version represents the next iteration of technology acquired through DeepHealth’s recent acquisition of French AI firm Gleamer for up to $270 million. The software is part of DeepHealth’s Radiology AI Suite, which includes clinical AI and workflow applications spanning breast, neurological, thoracic, musculoskeletal, thyroid, prostate, abdominal and pelvic imaging.
“This clearance advances our vision for an intelligent experience that helps move patients more efficiently from exam to informed action,” said Niccolò Stefani, MD, business and product leader, clinical AI, at DeepHealth. “Chest XRay supports radiologists across the volume of everyday interpretation and the urgency of high-impact acute care, bringing clinically relevant information forward when and where it can help shape the next step of a patient’s care journey.”
Chest XRay is now commercially available in the U.S. and Europe, and existing customers are eligible to receive the updated version.
Read more here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.