Meet Clairity at RSNA 2025
Clairity Breast is the first and only FDA De Novo–authorized, image-only AI model that predicts a woman’s 5-year breast cancer risk from an existing screening mammogram.
Meet Clairity at RSNA 2025
CLAIRITY BREAST transforms routine mammograms into forward-looking clinical insight. Join us at RSNA 2025 to explore how it unlocks new value for your clinical programs.
Set the standard with predictive AI for breast health
At RSNA 2025, Clairity will present groundbreaking studies that reveal how image-only AI models can expand the role of mammography from detection to risk prediction.
Explore Our PresentationS
Clairity at RSNA 2025
What’s New and Why It Matters
Presented by Clairity’s Founder,
Dr. Constance Lehman, MD, PhD
Nov 30 2025 at 11:45AM CT
Session (S3A-SPBR-1): Predicting future risk of DCIS and invasive breast cancer using AI: A multi-site study in U.S. and European screening
- Summary: The presentation explores how AI predicts both invasive cancer and DCIS, providing a fuller picture of risk than traditional tools.
- Abstract can be viewed here.
Nov 30 2025 at 2:30PM CT
Session (S5-SSBR02-5): External validation of discrimination and calibration of an image-only AI model for 5-year breast cancer risk prediction across geographically diverse screening sites
- Summary: Attend this session to learn how an image-only AI model, validated on more than 122,000 mammograms from 10 diverse U.S. health systems, can predict 5-year breast cancer risk with strong accuracy and reliable calibration.
- Abstract can be viewed here.
Nov 30 2025 at 2:30PM CT
Session (S5-SSBR02-2): Image-only AI model compared with breast density for identifying women at 5-year increased risk of breast cancer: multicenter U.S.-European analysis
- Summary: Learn how a deep learning model, trained on more than 400,000 mammograms, provides stronger and more precise risk stratification than breast density alone.
- Abstract can be viewed here.
Dec 2 2025 at 12:45PM CT
Session (T5B-SPBR-1): Targeted risk stratification in women under 50 using an image-only AI model: addressing the rise in breast cancer incidence in younger women
- Summary: Join this session at RSNA 2025 to learn how AI is uncovering hidden risk in women under 50 and what this means for the future of personalized screening.
- Abstract can be viewed here.
Dec 2 2025 at 3:15PM CT
Session (T7-STCE2-1): Unlocking the full value of the mammogram: using AI-based risk scores to detect cancer, identify missed cancers, and predict future risk
- Summary: This study evaluated an AI-based image-only risk model applied to routine screening mammograms, showing it can detect current cancers, flag missed cancers, and predict future breast cancer risk years in advance.
- Abstract can be viewed here.
Special Research
Dec 1 2025 at 3:00PM CT
Session (M7-SSBR03): Do deep learning breast cancer risk scores change meaningfully over time? A longitudinal analysis
- Summary: In addition, we highlight a presentation in the domain of image-only AI risk scores, showing the scores are dynamic, not static, and rise steadily over time in women who go on to develop cancer, while remaining flat in women who stay cancer-free. This finding underscores a potential paradigm shift: breast cancer risk is not a one-time calculation but a trajectory that can be tracked to guide earlier, more personalized interventions.
- Abstract can be viewed here. In addition to her work leading research on Clairity Breast, Dr. Lehman also conducts studies in the broader field of image-based AI, including the Mirai model.
About Dr. Constance Lehman
Constance “Connie” Lehman, MD, PhD, is Professor of Radiology at Harvard Medical School, founder of Clairity, and co-Director of the Breast Imaging Research Center at Massachusetts General Hospital. A graduate of Duke University with MD and PhD degrees from Yale, Dr. Lehman has led pioneering research demonstrating how AI can surpass traditional methods in predicting breast cancer risk and interpreting mammograms.
The First FDA-Authorized
Image-Only Risk Model
Clairity Breast is the first and only FDA De Novo–authorized AI platform that predicts a woman’s 5-year risk of developing breast cancer directly from her screening mammogram.
Developed using more than 1.7 million mammograms and externally validated in studies involving over 122,000 exams across 10 U.S. health systems, the model delivers accurate, well-calibrated risk estimates across age, race, and breast density.
No questionnaires, no added imaging, and no workflow changes. Clairity Breast integrates seamlessly into routine screening programs and supports more precise, equitable breast cancer prevention.
FDA De Novo–Authorized AI Risk Model
Mammograms Used in the Development of Clairity Breast
Screening Mammograms Used in External Validation
Built for enterprise.
Why This Matters for Clinical Practice
Accurately identifies women at high risk, including those missed by traditional models like Gail or Tyrer-Cuzick.
AI-identified high-risk women have significantly higher cancer detection rates on MRI, enabling better use of limited resources.
Validated across diverse populations, showing consistent accuracy across race, age, and breast density.
Supports NCCN-consistent follow-up pathways, including MRI, ultrasound, genetic testing, and risk-reduction counseling.
In the News
Published In & Presented At
Clairity Breast builds on a strong foundation of peer-reviewed research in image-only AI risk prediction. Our FDA-authorized model has now been validated across 122,000+ screening mammograms and is being presented at RSNA 2025.
