Intended Use

Intended to be used by endoscopists during video colonoscopy to assist identification of lesions such as polyps by highlighting regions with visual characteristics consistent with mucosal abnormalities. It is an adjunct to standard white-light endoscopy and not intended for diagnosis or to replace histopathological sampling.

Technology

The system acquires digital video output from an endoscopy camera, performs real-time processing using deep machine learning (deep neural networks) to detect regions consistent with polyps, and outputs annotated video indicating suspicious areas on a touchscreen and additional monitors. Major elements include a cart, computing device, touchscreen, and HDMI splitter with software for real-time AI detection of lesions in white-light endoscopy videos.

Performance

Performance was validated offline on 172 colonoscopy videos containing 449 polyps with excellent polyp-wise recall (up to 100% detection for small to large polyps). Clinical testing in a randomized multi-center trial with 950 patients showed that ME-APDS-assisted colonoscopy detected 37% more adenomas per colonoscopy than conventional colonoscopy, with no increase in adverse events or clinically relevant procedure delays. The system showed similar safety and effectiveness compared to the predicate GI Genius device.

Predicate Devices

No predicate devices specified

Device Timeline

  • 1

    Submission

    11/17/2022

    8 months
  • 2

    FDA Approval

    7/25/2023

Ready to Sharpen Your Edge?

Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.