Deep Recon is an AI-powered image reconstruction software designed for CT scanners. It uses deep learning technology to produce clearer CT images with less noise and better low contrast detectability, helping clinicians get accurate diagnostic images while potentially reducing the radiation dose for patients. It supports scans of the head, chest, abdomen, cardiac, and vascular regions, and is integrated into specific United Imaging CT scanners.
Intended Use
Deep Recon is a data driven image reconstruction method based on deep learning technology. It is intended to produce cross-sectional images by computer reconstruction of X-ray transmission data taken at different angles planes, including Axial, Helical, and Cardiac acquisition. Deep Recon is designed to generate CT images with lower image noise, and improved low contrast detectability, and it can reduce the dose required for diagnostic CT imaging. It can be used for head, chest, abdomen, cardiac and vascular CT applications for adults and is intended to be used with uCT 760 and uCT 780 only.
Technology
Deep Recon uses dedicated deep neural networks trained on low dose filtered back projection (FBP) images from UIH's CT scanners uCT 760 and uCT 780. It is integrated as part of the CT reconstruction chain to produce images resembling traditional FBP but with reduced noise and improved low contrast detectability. The user selects reconstruction type, convolution kernel, and strength (noise index level).
Performance
Non-clinical testing included image performance tests on phantoms for low contrast detectability, image noise, mean CT number, uniformity, spatial resolution, and reconstructed section thickness, showing that Deep Recon had equivalent or better performance than Filtered Back Projection. Clinical image evaluation was performed on retrospective case studies scored by board-certified radiologists, demonstrating that Deep Recon images are equivalent or better in diagnostic quality. Additional studies showed low dose images with Deep Recon are equivalent or better than standard dose images with FBP. No prospective clinical study was conducted.
Predicate Devices
No predicate devices specified
Device Timeline
- 1
Submission
11/4/2019
8 months - 2
FDA Approval
7/6/2020
Other devices from Shanghai United Imaging Healthcare Co., Ltd.
- uDR 780i2024 • 510K
- uDR 380i Pro2024 • 510K
- uMI Panorama2024 • 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.