According to the
International Agency for Research on Cancer, in 2018 breast cancer accounted
for 26.4 percent of all new cancer cases in European women.[i] Europe’s
radiologist shortage, particularly in Scotland and the United Kingdom, where the
shortage is dire, makes it hard to maintain medical standards.[ii] Without
a sufficient number of radiologists for 1st and 2nd
reads, there’s a risk that abnormalities may be missed.
The close
scrutiny radiologists give to each image takes time and can be visually and
mentally exhausting. Deep-learning, artificial intelligence (AI) technology can
assist in breast cancer detection and decision-making. ProFound AI for 2D
Mammography, a deep-learning AI platform trained with one of the largest
available image datasets, can improve diagnostic confidence.
Professor Sylvia H. Heywang-Köbrunner,
a radiologist and researcher based in Munich, Germany, was scheduled to present
her findings at the 2020 European College of Radiology (ECR) annual conference
in March.[iii] In this
study, 18,002 consecutive screening mammograms acquired between January and
November 2018 were anonymized and processed using ProFound AI for 2D
Mammography. The data show that using a case threshold of 30, Profound AI for
2D Mammography achieved a sensitivity of 91.5 percent and a specificity of 80.2
percent for 32 ductal carcinoma in situ (DCIS) and 85 invasive cancers. The
first reader’s results were 84.6 percent for sensitivity and 91.6 percent for
specificity, while the second reader’s results were 89.7 percent and 91.5
percent, respectively. These results validate the efficacy of ProFound AI for
2D Mammography. In her conclusion, Professor Heywang-Köbrunner wrote, “Achieved
results justify hopes to use novel CAD systems for a second (e.g., in countries
with a shortage of readers) or third reading in the near future. Human
consensus reading remains indispensable.”
In addition, when
using findings from ProFound AI for 2D Mammography, radiologists can reduce the
likelihood of missing malignancies and decrease the number of false positives. “With
such a high sensitivity and specificity, ProFound AI often reveals suspicious
areas that may have been missed, and at the same time, reassures my decisions
when no further patient procedures are needed,” said Axel Gräwingholt, MD,
Radiologie am Theater, Paderborn, Germany.
Profound AI for
2D Mammography rapidly and accurately analyzes each individual image or slice,
detecting both malignant soft tissue densities and calcifications with accuracy.
It can be particularly useful for detection in dense breast cases and for
complex cases with subtle lesions. “The arrival of ProFound AI marks a
real technological advancement in the development of breast cancer detection
tools. I use ProFound AI in my daily practice, and it helps me in my diagnostic
decision-making,” said Corinne Balleyguier, MD, PhD, radiologist at Gustave
Roussy Cancer Campus, Villejuif, France. “In the beginning, I was skeptical
about how AI could help me, but I quickly understood the assistance ProFound AI
could bring, especially with subtle lesions and complex cases.”
ProFound AI for
2D Mammography highlights suspicious areas and assigns unique Certainty of
Finding and Case Scores that radiologists can use to make clinical decisions
and prioritize caseloads. ProFound AI for 2D Mammography’s detection
capabilities improve continuously through ongoing software updates. Radiologists
can use the AI platform to prioritize their caseloads and manage their
workflow.
In 2019, Profound
AI for 2D Mammography was featured at several medical conferences, including the
Société Française d’Imagerie de la FEMme (SIFEM), the Journées Francophones de
Radiologie (JFR) Congress in Paris and the European Society of Breast Imaging
(EUSOBI) Annual Scientific Meeting in Budapest. In July 2019, ProFound AI for
2D Mammography received CE Mark approval. This approval signifies that it
conforms to stringent European standards for health, safety and environmental
protection, and further validates the product as a world-class AI solution.
Early detection
is key for improving health outcomes and longer-term survival odds. Finding
tumors earlier means they’re easier to treat. ProFound AI for 2D Mammography can
reduce the number of false-positives and callbacks and help make the entire
process less stressful for patients.
Part of iCAD’s deep-learning,
artificial intelligence platform, ProFound AI also includes Digital Breast
Tomosynthesis (DBT), which was CE Marked in March 2018 and FDA cleared in
December 2018.
[i]
World Health Organization. Globocan 2018. Accessed via https://gco.iarc.fr/today/data/factsheets/populations/908-europe-fact-sheets.pdf
[ii]
Silvestrin, A. Europe’s Looming Radiology Capacity Challenge: A Comparitive
Study. (2016). Accessed via https://www.telemedicineclinic.com/wp-content/uploads/2016/11/Europes_looming_radiology_capacity_challenge-A_comparitive_study.pdf
[iii] The
value of 2D-AI-based CAD for second or third reading tested on 17,910 screening
mammograms [RPS 702-4]. https://event.crowdcompass.com/ecr2020/activity/78pY0IUG4N