Skip to content

In recent weeks, the
number of people diagnosed with the coronavirus, or COVID-19, has rapidly
increased, impacting our lives in unprecedented ways and placing a growing
burden upon hospitals and clinicians worldwide. This emerging issue has led organizations
such as the American College of Radiology, the American Cancer Society, and
Susan G. Komen to advise lower-risk women to consider rescheduling routine
mammograms so hospitals may address critical COVID-19 cases. [i],[ii],[iii] However,
this guidance may be confusing to both women and clinicians, and may cause some
patients with more urgent and higher risk factors to be deferred, which could
ultimately have negative long-term consequences for them and the healthcare
system as a whole in the long run. Additionally, once the threat of COVID-19
passes, radiologists may be faced with a significant backlog of women whose
breast cancer screenings are overdue.

ProFound AI™ can help
radiologists enhance patient care and address these issues during these challenging
times. As the first artificial intelligence (AI) software for digital breast
tomosynthesis (DBT) to be FDA-cleared, this high-performing workflow solution can
help radiologists assess which patients should not delay screening and presents
a solution to the growing mammography backlog that will need to be cleared once
the threat of COVID-19 passes.

Assessing Which Patients
Should Not Delay Screening

“Studies from published
academic research show that on average, a mammography provider can expect to
detect up to 6 cancers per 1,000 screening mammograms,[iv] which
indicates that even in the face of this pandemic, cancer is not going away,”
according to Stacey Stevens, President of iCAD. “Now more than ever, it is
increasingly critical to empower radiologists with technology such as ProFound
AI, which can help them quickly and effectively decide which patients should
not defer mammography screening.”

ProFound AI is a
high-performing workflow solution for 2D and 3D mammography, or digital breast
tomosynthesis (DBT), featuring the latest in deep-learning artificial
intelligence (AI). Trained with one of the largest available DBT datasets,
ProFound AI rapidly and accurately analyzes each DBT image, or slice, and provides
radiologists with key information that assists radiologists in clinical
decision-making and improving reading efficiency. The technology offers
clinically proven benefits to patients and clinicians, including an 8 percent
improvement in sensitivity and a 7.2 percent reduction in the rate of false
positives and unnecessary recalls,[v] and also provides crucial data, such
as Case Scores, which can help radiologists evaluate which patients should not
delay screening.

“In the face of COVID-19,
improving accuracy and efficiency for breast cancer screening is now more
essential than ever. Using this cutting-edge technology, radiologists have the
ability to review prior year mammograms and gauge which patients should not
delay breast cancer screening based on the technology’s unique Case Score,
which represents a relative level of suspicion for the case containing
potentially cancerous findings that may require further workup,” according to
Mark Traill, MD, radiologist at University of Michigan Health, Metro Health.
“It is clinically proven to help radiologists significantly increase their
productivity. ProFound AI can also reduce the rate of false positives,1
which are not only stressful for patients, but also place a significant
additional burden on providers. Clinicians using ProFound AI will be very happy
they have such a sophisticated tool helping them better position their
screening programs’ recovery from this unprecedented global disruption.”

Managing the Backlog

Nearly 40 million
mammograms are performed in the U.S. annually,[vi] which
translates to an average of 3.3 million women screened per month. If some of
these women defer screening until later this year, radiologists could suddenly be
faced with surge of patients requiring mammograms when COVID-19 subsides. Additionally,
in recent years DBT has rapidly grown in adoption in the U.S. providing
radiologists with more information than ever before, but also increasing
reading time substantially. While traditional 2D mammography yields four images
per patient, DBT produces hundreds of images per
patient, which is an exponential increase in the number of images radiologists
must review each day.

“Physicians and the
healthcare system as a whole will be faced with an enormous backlog of
mammograms when the threat of COVID-19 passes. ProFound AI is uniquely
positioned to address this emerging challenge,” according to Michael Klein,
Chairman and CEO of iCAD.

ProFound AI offers
clinically proven time-savings benefits for clinicians, which can help them
accurately and efficiently review cases in less time. According to results from
a large reader study, the technology was found to cut reading time for
radiologists reading DBT by 52.7 percent.v For cases with dense breasts, which can be
particularly challenging for radiologists to read, ProFound AI for DBT reduced
reading time by up to 57.4 percent.[vii] Additionally,
ProFound AI’s unique Case Scores provide radiologists with key information that
can help them choose which patients need to be screened first, which may help
to address the growing logjam of mammograms due to COVID-19.

As the field of
mammography moves from age-based screening protocols to more personalized
risk-adaptive screening, assessing which patients should be screened sooner can
enhance patient care at your radiology practice, both during and after the
COVID-19 pandemic. We are proud to offer this world-class solution and empower our
customers to be on the front-line of this exciting new realm, which will truly
have a positive impact on patient care and the healthcare system.


[i] American College of
Radiology (ACR). COVID-19 Radiology-Specific Resources. Accessed via



[iv] Conant, E. et al.
(2020). Five Consecutive Years of Screening with Digital Breast Tomosynthesis:
Outcomes by Screening Year and Round. Radiology. Accessed via

[v] Conant, E et al.
(2019). Improving Accuracy and Efficiency with Concurrent Use of Artificial
Intelligence for Digital Breast Tomosynthesis. Radiology: Artificial
Intelligence. 1 (4). Accessed via

[vi] U.S. Food and Drug
Administration. Good News for Public Health: Most Mammography Facilities are in
Full Compliance with MQSA Regulations. Accessed via

[vii] Hoffmeister, J.
(2018). Artificial Intelligence for Digital Breast Tomosynthesis – Reader Study
Results. [White paper]. Accessed via