Clarify DL

Clarify DL is a software-only image reconstruction method for nuclear medicine bone SPECT scintigraphy that integrates a deep learning model to reduce noise and improve image quality while maintaining contrast. It is deployed within GE HealthCare's Xeleris V Processing and Review System and StarGuide devices and may be used for patients of all ages.

Clarify DL adds an integrated deep learning model trained to reduce noise to the predicate devices' OSEM-based (and BSREM-based for StarGuide) iterative image reconstruction. It utilizes the same hardware and software platform as the predicates and produces substantially equivalent images with improved contrast-to-noise ratio and contrast recovery coefficient.

The device was designed under Quality System Regulations 21 CFR 820 and ISO 13485. Testing included Structure Similarity Index Measure (SSIM), Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Contrast-to-Noise Ratio (CNR), and Contrast Recovery Coefficient (CRC) metrics.

Clarify DL is substantially equivalent because it does not change the indications for use, utilizes the same hardware and software platform as the predicates, and produces images of equivalent diagnostic quality. Engineering bench testing and clinical reader evaluation by expert nuclear medicine physicians demonstrated that image quality, resolution, and noise characteristics are comparable to predicate devices, with no new safety or effectiveness questions identified.