GE HealthCare (GEHC) has received premarket approval from the U.S. Food and Drug Administration (FDA) for Pristina Recon DL, 3D mammography image reconstruction software powered by deep learning (DL).
Pristina Recon DL leverages two DL models: the first reconstructs high-fidelity 3D volumes while minimizing artifacts and perceived noise, and the second is trained to enhance the visualization of clinically relevant information in the DL synthesized 2D view, GEHC said.
The technology is an enhancement to its Pristina Via system and is the first in mammography to use deep learning in combination with iterative reconstruction to provide digital breast tomosynthesis (DBT) image quality without compromising on patient dose, according to the firm. Pristina Via with Recon DL utilizes Nvidia RTX accelerated computing technology to execute its image reconstruction, which delivers fast and accurate images in the exam room and for clinical diagnosis, GE HealthCare added.
In other GEHC news, the company plans to launch Genesis Radiology Workspace, a cloud-based system for optimizing radiology workflows anchored by a new viewer called Genesis View.
Genesis View is a zero-footprint viewer designed to allow users to work from virtually anywhere with diagnostic accuracy and provides 2D/3D visualization and AI tools, according to the firm. The system will also integrate with GEHC's Genesis portfolio, its cloud-enabled software as a service (SaaS) tool for enterprise imaging.
The system is designed to adapt to individual user reading preferences and automatically display current and prior studies, which can help save time and reduce cognitive load, the company said. It allows user-defined AI prioritization so that critical findings can be brought to the top to make sure they are visible, and enables radiologists to reject or modify AI findings provided by third-party apps based on their review.
Genesis View is pending 510(k) clearance by the FDA, the company noted.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&dpr=2&fit=crop&h=167&q=70&w=250)
