New Mammography Studies Assess Image-Based AI Risk Models and Breast Arterial Calcification Detection
In a recent interview, Emily Conant, M.D., discussed findings from two studies presented at the European Congress of Radiology (ECR) conference that showed the potential impact of image-based AI risk models and breast arterial calcification detection in mammography screening.
The quantitative assessments provided by image-based artificial intelligence (AI) models in mammography screening may have a significant impact in ascertaining short-term breast cancer risk, according to new research presented at the European Congress of Radiology (ECR) conference.
In a study involving over 8,000 women (including 780 women with incident breast cancer), researchers found that AI-based risk models (ProFound AI Risk, iCAD) for digital mammography and digital breast tomosynthesis (DBT) predicted 40 percent and 48 percent, respectively, of future breast cancers in high-risk women.
“Another important thing is this (research was focused on) short-term risk, so this is really important to guide the choice of precision screening for women. Who needs what? How often? … How can we really use this information, this risk based on the image qualities, to drive precision screening better than we're doing it right now?,” posited lead study author Emily Conant, M.D., a professor emeritus in the Department of Radiology at the Hospital of the University of Pennsylvania.
In a recent interview, Dr. Conant also discussed another study presented at the ECR conference that examined the use of an algorithm to facilitate detection of breast arterial calcification (BAC) in mammograms. Dr. Conant and her colleagues found that the algorithm achieved a 98 percent case-level AUC for BAC detection.
“I think the fact that the mammogram shows more than just, you know, breast cancer and risk for future breast cancer, short-term risk, long-term risk, but can also help sort of triage women who are at higher cardiac risk. That's really huge,” maintained Dr. Conant. “Again, we need more studies, we need more data, we need prospective studies and all of that. But this is a rapidly expanding area that I think really adds benefit and worth, etc, to mammographic screening.”
(Editor’s note: For related content, see “Study: Mammography AI Leads to 29 Percent Increase in Breast Cancer Detection,” “Mammography Study Shows Merits of AI for Improving Breast Cancer Detection and Effectiveness of Recalls” and “Study Shows Link Between Breast Arterial Calcification and Significant Cardiovascular Risks.”)
For more insights from Dr. Conant, watch the video below.
New Collaboration Offers Promise of Automating Prior Authorizations in Radiology with AI
March 26th 2025In addition to a variety of tools to promote radiology workflow efficiencies, the integration of the Gravity AI tools into the PowerServer RIS platform may reduce time-consuming prior authorizations to minutes for completion.
Predicting Diabetes on CT Scans: What New Research Reveals with Pancreatic Imaging Biomarkers
March 25th 2025Attenuation-based biomarkers on computed tomography (CT) scans demonstrated a 93 percent interclass correlation coefficient (ICC) agreement across three pancreatic segmentation algorithms for predicting diabetes, according to a study involving over 9,700 patients.
AI-Initiated Recalls After Screening Mammography Demonstrate Higher PPV for Breast Cancer
March 18th 2025While recalls initiated by one of two reviewing radiologists after screening mammography were nearly 10 percent higher than recalls initiated by an AI software, the AI-initiated recalls had an 85 percent higher positive predictive value for breast cancer, according to a new study.
