Deep Learning Software for 3D Mammography Reconstruction Gets FDA Premarket Authorization

The Food and Drug Administration (FDA) has granted premarket authorization for Pristina Recon DL, which is reportedly the first combination of deep learning with iterative reconstruction capabilities geared toward bolstering image quality for digital breast tomosynthesis (DBT).

Facilitating artifact and noise reduction, the Pristina Recon DL software offers reconstruction of high-fidelity 3D volumes along with deep learning-enhanced visualization for synthesized 2D mammography images, according to GE HealthCare, the developer of Pristina Recon DL.

The company noted that recent research with the Pristina Recon DL software revealed superior detection of microcalcification clusters and masses.

“With Pristina Via with Recon DL, we’re setting a new benchmark in breast imaging — delivering sharper, clearer, and more consistent images¹⁴ that empower radiologists with more confidence,” said Pooja Pathak, the vice president and general manager of mammography at GE HealthCare. “As an upgradable feature on the Pristina Via platform, we are excited to now offer customers uncompromised image quality combined with fast, accurate workflows.”