FDA approves GE HealthCare’s AI-powered mammography reconstruction technology

USA— The US Food and Drug Administration (FDA) has granted premarket authorization (PMA) to GE HealthCare’s Pristina Recon DL tool, marking a significant advancement in breast imaging technology.

Recon DL represents a 3D mammography image reconstruction technology that functions as an upgradeable feature on GE HealthCare’s Pristina Via platform.

The system employs two deep learning models working together to distinguish meaningful signals from background noise in mammography images, thereby enhancing the clarity and diagnostic value of breast scans.

GE HealthCare positions Recon DL as the first mammography technology to combine deep learning with iterative reconstruction methods to enhance digital breast tomosynthesis (DBT) image quality.

This dual approach addresses longstanding challenges in breast imaging by improving both the technical quality of images and their clinical utility.

The technology operates through a two-stage process.

The first deep learning model reconstructs 3D volumes while minimizing artifacts that can obscure important details.

Meanwhile, the second model focuses on improving the visualization of clinically relevant information, which it then incorporates into a synthesized 2D view that radiologists can easily interpret.

A clinical study conducted by GE HealthCare to support the PMA submission demonstrated impressive results.

Breast radiologists preferred Recon DL’s overall image quality in 99.1% of image reviews when compared to previous DBT reconstruction tools.

Additionally, the tool showed strong performance in detecting microcalcification clusters and masses, as evidenced in a trial using modeled clinical data.

Throughout the development process, GE HealthCare collaborated with academic institutions and imaging centers to develop and validate the algorithms powering Recon DL.

This collaborative approach helped ensure the technology meets real-world clinical needs and standards.

Jyoti Gupta, president and CEO of GE HealthCare’s women’s health and X-ray division, emphasized the significance of this advancement.

 She stated that applying advanced deep learning technologies shapes the future of breast imaging through uncompromised image quality, faster workflows, and greater confidence in early cancer detection.

The timing of this approval carries particular importance given the prevalence of breast cancer.

Research from the National Cancer Institute (NCI) shows that breast cancer remains one of the most common cancers among women, with one in eight women expected to receive a diagnosis during their lifetime.

The mammography equipment market continues to expand significantly.

According to GlobalData analysis, this market segment, which falls under the broader diagnostic imaging category, grows at a CAGR of 4.9%.

The market valuation stands at USD2.4 billion in 2024 and forecasts project it will reach nearly USD4 billion by 2034.

GE HealthCare faces competition in this advancing field.

In January, Siemens Healthineers began installing its Mammomat B.brilliant, a mammography system that also incorporates 3D image reconstruction technology.