As synthetic mammography keeps improving, it may soon be the new standard
Full-field digital mammography (FFDM) is considered the gold standard for 2D breast cancer screening due to its consistency, cost-effectiveness and wide availability.
The U.S. Food and Drug Administration (FDA) has continued to clear alternative techniques, however, including synthetic mammography, which can offer key advantages to patient health when combined with digital breast tomosynthesis (DBT).
Synthetic mammography is an image reconstruction technique, artificially creating a 2D image based on existing DBT data sets. But, how accurate is the technique, and what are the pros and cons of synthetic mammography? Researchers from Brigham and Women's Hospital in Boston have attempted to answer these questions with a new analysis in RadioGraphics.[1]
As emphasized throughout the paper, it’s the reduced radiation exposure that is the primary selling point for synthetic mammography. While DBT produces 3D images of the breast, it requires a 2D image to complete the puzzle – and that can often lead to FFDM and additional radiation exposure. As noted, synthetic mammography artificially creates the necessary 2D image, limiting radiation. But it does have its own drawbacks in terms of reliability: lower spatial resolution, reduced visibility and a higher rate of false-positive results.
There is also a risk of more image artifacts with synthetic mammography, though it seems an awareness of how they form, along with properly examining the artifacts, can help to mitigate the problem:
“Although motion artifacts are theoretically fewer with synthetic mammography than with FFDM due to the decreased acquisition time, the evaluation of motion artifacts on synthetic mammography images can be challenging,” wrote lead author Sona Chikarmane, MD, a radiologist with Brigham and Women's Hospital, and colleagues. “Careful evaluation of the skin folds, wavy metallic artifacts, and the DBT cine clip of the inframammary fold will allow evaluation of possible motion. Given the inherent decreased spatial resolution of synthetic mammography, as compared with that of FFDM, it is of utmost importance to evaluate for motion artifacts to further avoid missing subtle findings such as fine and/or amorphous microcalcifications.”
Synthetic mammography as the new standard
In their literature review, the researchers noted that, despite its limitations, synthetic mammography is set to overtake FFDM when used with DBT. In part, that is because the lower radiation is great for breast cancer patients – especially those facing surgery. And as synthetic mammography becomes more popular and accessible, the underlying technology is improving – and advancements in AI may improve it further.
“The application of artificial intelligence has been used to improve the quality of synthetic mammography image reconstruction. Machine learning can be applied to enhance suspicious findings and remove normal tissue, which may obscure lesions,” the authors added. “Computer-aided diagnosis to enhance synthetic mammography has been shown to lead to improved diagnostic performance by automatically identifying suspicious lesions on each DBT slice and visually enhancing these findings with synthetic mammography.”
According to the authors, companies are working on updated versions of synthetic mammography algorithms aimed at producing sharper, more natural images. With investment from companies and regulatory approval, improvement on synthetic mammography is only a matter of time – and If the technology can be properly tweaked, the benefits may well outweigh the disadvantages.
Chad is an award-winning writer and editor with over 15 years of experience working in media. He has a decade-long professional background in healthcare, working as a writer and in public relations.