Pristina Recon DL tool from GE HealthCare which uses two deep learning -DL models has gone on to receive a pre-market authorisation – PMA from the US Food and Drug Administration – FDA.
Recon DL is a 3D mammography image reconstruction technology that is available as an upgradeable feature on the Pristina Via platform from GE HealthCare. It makes use of two deep learning -DL models in tandem in order to separate meaningful signals coming from noise in the mammography images.
Recon DL tool from GE HealthCare is the first mammography technology to make use of deep learning in combination with iterative reconstruction in order to improve the digital breast tomosynthesis – DBT image quality.
The first DL model goes on to reconstruct the 3D volumes in order to minimize the artifacts, whereas the second is trained to enhance the visualization of certain clinically relevant information in a synthesized 2D view.
A GE HealthCare study that supported the PMA submission with the FDA went on to show that Recon DL’s complete image quality was indeed preferred by breast radiologists in 99.1% of the image reviews as compared to the previous DBT reconstruction tools.
The tool also went ahead and performed strongly when it comes to the detection of microcalcification clusters as well as masses, as demonstrated by a trial that modelled clinical data.
Jyoti Gupta, the women’s health and X-ray president and CEO at GE HealthCare, commented that through applying advanced deep learning tech, they are indeed shaping the future when it comes to breast imaging – the one that is defined through faster workflows, uncompromised image quality, and greater confidence in terms of cancer detection.
GE HealthCare confirmed that it has worked along with the academic institutions as well as imaging centers in order to develop as well as validate the algorithms that are used in Recon DL all across its development process.
National Cancer Institute – NCI research indicates that breast cancer still remains one of the most common cancers in women, with one in every eight expected to go through a diagnosis in a lifetime.
GlobalData analysis says that the global mammography equipment market, which is a subcategory of the broader diagnostic imaging market, is witnessing a surge at a CAGR of 4.9% and also forecasts to reach a valuation of somewhere around $4bn by 2034, which is up from the $2.4bn that was seen in 2024.
Apparently, in January 2024, Siemens Healthineers started the first installations of its Mammomat B.brilliant, which is a mammography system also featuring 3D image reconstruction technology.
GE HealthCare secures PMA for deep-learning 3D mammography system, which was created and published originally by a GlobalData-owned brand – Medical Device Network.