GE HealthCare Receives FDA Authorization for AI-Powered 3D Mammography Technology
Key Insights
GE HealthCare received FDA Premarket Authorization for Pristina Recon DL (search), the first mammography technology to combine deep learning with iterative reconstruction for enhanced 3D breast imaging.
A clinical study demonstrated that breast radiologists preferred Pristina Recon DL (search) image quality in 99.1% of reviews compared to previous digital breast tomosynthesis reconstruction methods.
The technology utilizes two sequential deep learning models to minimize artifacts and enhance visualization of clinically relevant information while maintaining patient dose safety.
GE HealthCare has received FDA Premarket Authorization for Pristina Recon DL (search), a breakthrough deep learning-based image reconstruction technology for 3D mammography that promises to enhance breast cancer (search) detection capabilities. The technology represents the first mammography system to combine deep learning with iterative reconstruction for digital breast tomosynthesis image quality.
Advanced AI Architecture Delivers Superior Image Quality
Pristina Recon DL (search) leverages two deep learning models working in sequence to separate meaningful signal from noise. The first model reconstructs high fidelity 3D volumes with greater purity, minimizing artifacts and perceived noise. The second model enhances visualization of clinically relevant information in the synthesized 2D view.
The system utilizes NVIDIA (search) RTX accelerated computing technology for image reconstruction, delivering fast and accurate images in the exam room and for clinical diagnosis. This enhancement to GE HealthCare's Pristina Via system maintains outstanding digital breast tomosynthesis image quality without compromising patient dose.
Clinical Validation Shows Strong Radiologist Preference
A recent study demonstrated compelling clinical performance, with Pristina Recon DL (search) overall image quality preferred by breast radiologists in 99.1% of image reviews compared to previous digital breast tomosynthesis reconstruction. The study was performed on 140 cases with 8 MQSA-approved radiologists trained on clinical image quality by the American College of Radiology (search).
The technology also offers superior performance in the detection of microcalcification clusters and masses, as demonstrated in a trial with modeled clinical data.
Addressing Critical Healthcare Need
Breast cancer (search) remains one of the most common cancers among women, with one in eight expected to face a diagnosis in their lifetime, according to the National Cancer Institute. An estimated 1.1 million breast cancer-related deaths are projected annually by 2050, highlighting the critical need for advanced detection technologies.
"By applying advanced deep learning technologies, we're shaping the future of breast imaging – one defined by uncompromised image quality, faster workflows, and greater confidence in early cancer detection," said Jyoti Gupta, PhD, President and CEO, Women's Health and X-ray at GE HealthCare.
Enhanced Workflow and System Performance
The Pristina Via with Recon DL system offers significant operational improvements including zero-click acquisition within protocols and zero wait time with 2.3 seconds elapsed time between end of exposure and system ready in digital breast tomosynthesis. The system demonstrates 99% uptime based on analysis of 42,543 service requests from October 2023 to September 2024.
Additional features include a 15-minute biopsy exam with Serena add-on device and Contrast Enhanced Mammography with SenoBright HD. Pristina Recon DL (search) is available as an upgradable feature on the Pristina Via platform.
Industry Partnership and Development
GE HealthCare developed and validated the algorithms with academic institutions and high-volume outpatient imaging centers throughout the development process. The collaboration approach ensured real-world validation of the technology's clinical utility.
"Our collaboration with GE HealthCare has been instrumental in advancing breast imaging capabilities, and the new 3D image quality represents a meaningful upgrade that will benefit radiologists and patients alike," said Dr. Howard Berger, President and Chief Executive Officer of RadNet. "This pioneering AI technology will help elevate breast care by delivering the clarity and consistency radiologists need to enable more confident diagnoses."