SIGNA Premier

The SIGNA Premier MRI system is enhanced with AIR Recon DL, a novel software feature utilizing deep learning for image reconstruction. This technology aims to improve diagnostic image quality through increased signal-to-noise ratio and sharpness, while also enabling reduced scan times without compromising these critical image characteristics.

The SIGNA Premier is a whole-body magnetic resonance imaging (MRI) system featuring a 3.0 Tesla superconducting magnet with a 70cm bore, designed for comprehensive diagnostic imaging across various anatomical regions. This system incorporates a significant software enhancement called AIR Recon DL, a deep-learning based reconstruction technique. The primary purpose of AIR Recon DL is to improve image quality by increasing signal-to-noise ratio (SNR) and enhancing image sharpness. Additionally, this software innovation aims to enable shorter scan times for patients while preserving the achieved SNR and image sharpness, thereby improving both diagnostic confidence and operational efficiency. The AIR Recon DL feature is integrated into the SIGNA Premier's existing software, requiring no hardware modifications, and provides operators with new options to select the reconstruction method and adjust the level of image noise reduction. Extensive nonclinical and clinical testing was conducted to validate the performance of the AIR Recon DL feature. Nonclinical tests using digital reference objects and phantom imaging demonstrated that AIR Recon DL successfully improves SNR and image sharpness, maintains low contrast detectability, and has minimal impact on noise spectral content or quantitative measurements, even with reduced scan times or in the presence of motion. Clinical evaluations, including objective measures from in vivo images and blinded reader studies with radiologists, confirmed these benefits across various pulse sequences, anatomies, and clinical use cases, including those involving contrast agents and diverse pathologies. Radiologists consistently preferred images reconstructed with AIR Recon DL, noting equivalent or superior image quality and maintained lesion conspicuity, even when comparing AIR Recon DL images from shorter scans to standard images from longer scans. The testing concluded that AIR Recon DL meets its product claims for improved image quality and efficiency.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific deep learning architecture or model is utilized in AIR Recon DL, and how was it trained (e.g., dataset size, diversity, validation methods)?; How does AIR Recon DL's performance vary across different magnetic field strengths or MRI systems, beyond the 3.0T SIGNA Premier?; Are there any specific types of artifacts that AIR Recon DL might be more prone to introducing or, conversely, particularly effective at mitigating compared to traditional reconstruction methods?; What is the quantifiable impact of AIR Recon DL on diagnostic accuracy or inter-reader variability in challenging clinical scenarios or for subtle pathologies?; Beyond SNR and sharpness, how does AIR Recon DL affect other quantitative imaging biomarkers (e.g., T1/T2 mapping, diffusion metrics) that rely on precise signal intensity values?