VIDEO: The rapid adoption of 3D mammography and use of AI to address dense breasts

 

Interview with Stamatia Destounis, MD, FACR, a radiologist and managing partner at Elizabeth Wende Breast Care in Rochester, New York, chair of the American College of Radiology (ACR) Breast Commission, and member of the Public Information Advisors Committee for Radiological Society of North America (RSNA) and Society of Breast Imaging (SBI) Communication Committee. She explains the rapid adoption of 3D mammogram digital breast tomosynthesis (DBT) technology as a better way to detect cancers in patients with dense breast tissue. She also explains how artificial intelligence (AI) will likely play a growing role in the coming years as a second set of eyes to detect cancers or other anomalies in the vastly larger and more complex DBT imaging studies. 

About 20 years ago, breast imaging rapidly moved from film analog mammography to full field digital mammography (FFDM) because it was a better technology. Today, film mammography systems are almost complete gone in the U.S. A similar trend is now being seen with DBT 3D mammography systems rapidly replacing FFDM systems. 

DBT systems are now approaching 50% of the accredited mammography systems installed in the U.S. While more expensive, with more images for the radiologists to reach, DBT does catch more cancers, reduces recall imaging and reduces the need for biopsies. For women with dense breasts, DBT had helped a lot in being able to see through the layers of tissue to better identify cancers. 

"We went from several years ago where very few facilities or hospitals had one unit, to now where we have facilities completely converting over to tomosynthesis. So the implementation has been pretty wide over a quick period of time," Destounis explained.

She said uptake was initially slow because there was no reimbursement for 3D mammograms, but that has changed, making it easier for centers to justify the expense of these systems. This is helping facilitate the transition away from digital mammography to tomosythesis. 

"There are several benefits to this, and the number one benefit is a reduction in call back rates, so woman are not getting called back for false alarms and false positives," Destounis explained. "That really makes life easier and reduces the workload for the radiologist so we don't have to focus so much time in calling patients."

The new mammography technology also offers better imaging.  

"We also have an increased cancer detection rate. With tomosynthesis we get a better mammogram where we can find more cancers and reduce patient anxiety. It is just a better test and I do thing it will continue to have wide implementation," Destounis said. 

DBT has limits and supplemental imaging is still needed

While DBT does help radiologists detect cancers, there are limits to 3D mammograms, and in some women with very dense breasts, supplemental imaging will still be necessary. 

She said tomo combined with breast ultrasound helps increase the number of cancers detected in these patients. In some states where breast MRI payments are covered, that is an even better modality. With advances in MRI technology, she said abbreviated breast MRI exams also can significantly reduce the exam times.

DBT now creates synthetic 2D mammograms

One of the issues with converting to 3D mammography was an issue with regulators over an apples to oranges comparison of prior 2D mammograms as compared to 3D datasets. This could require the patient to undergo a standard 2D mammograms and then a 3D exam, increasing the amount of radiation they received. This issue was solved by vendors programming the 3D systems to create a synthetic 2D mammogram view. 

Use of AI to speed longer reads on DBT

Digital breast tomo does take longer to read because the number of images in the exam increases from four to sometimes more than 40. It is more time consuming for radiologists to read these exams. However, she also believes AI will play a rising role to sort through the extra images to flag the ones radiologists should spend more time on, or even highlighting areas of interest. 

"Artificial intelligence is playing a role here, fine tuning areas of interest for us to look at," Destounis said. "There may be a role for AI to help radiologists, but we certainly are not saying AI will replace the breast radiologists. The early AI studies show AI can accurately predict risk for the patients based on the breast parenchymal patterns seen on the mammogram. I think there will be a role for AI to help act as a second set of eyes and to concentrate one these women that may be considered high risk based on their parenchymal pattern or some risk score that is assigned by AI." 
 

Dave Fornell is a digital editor with Cardiovascular Business and Radiology Business magazines. He has been covering healthcare for more than 16 years.

Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: dfornell@innovatehealthcare.com

Trimed Popup
Trimed Popup