Deep Learning Image Reconstruction
GE Healthcare Japan CorporationDeep Learning Image Reconstruction by GE Healthcare Japan Corporation is a deep learning based software integrated into CT scanners to reconstruct high-quality cross-sectional images of the head, whole body, cardiac, and vascular systems. It uses a trained deep neural network to reduce image noise and artifacts while maintaining spatial resolution, aiding clinicians in obtaining clearer diagnostic images with routine CT throughput.
Intended Use
The Deep Learning Image Reconstruction software is intended for head, whole body, cardiac, and vascular CT scans.
Technology
The software uses a dedicated deep neural network (DNN) trained specifically on CT data to model noise propagation and characteristics for noise reduction and artifact suppression. It integrates into the CT scanner's raw data image reconstruction pipeline and supports user-selectable reconstruction strength (Low, Medium, High). The technology is consistent with the predicate device and operates on GE's Edison platform.
Performance
Software underwent extensive design control and quality assurance testing including code review, software integration, safety testing, image performance verification, and simulating use validation. Bench testing compared low contrast detectability, noise, spatial resolution, streak artifact suppression, and other image quality metrics versus predicate devices using identical raw CT datasets. No new safety issues were identified.
Predicate Devices
No predicate devices specified
Device Timeline
- 1
Submission
3/23/2023
28 days - 2
FDA Approval
4/20/2023
Other devices from GE Healthcare Japan Corporation
- True Enhance DL2024 • 510K
- Deep Learning Image Reconstruction2022 • 510K
- Revolution Ascend2022 • 510K
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.