Using mammogram images and AI to improve breast cancer risk prediction
Official title: Advancing breast cancer risk prediction in national cohorts: the role of mammogram-based deep learning
Funded at Columbia University Health Sciences · NIH-11143084 · Data updated 2026-11-12
This project uses artificial intelligence on mammograms plus genetic risk information to make more accurate breast cancer risk predictions for women, with a focus on improving results for Black women.
Quick facts
| Grant type | U01 cooperative agreement |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Columbia University Health Sciences NIH-funded |
| Lab location | 1 site (New York, United States) |
| Project start | September 21, 2023 |
| Project end | August 31, 2026 |
| Funding year | 2025 |
| Award (this year) | $646,425 |
| NIH project number | 5U01CA276553-03 |
| Project ID | NIH-11143084 on NIH RePORTER |
Who could benefit from this research
✓Could help
The focus is on adult women who have screening mammograms and are represented in large U.S. cohorts—especially Black and non-Hispanic white women included in the Sister Study or the Black Women's Health Study.
✕Less likely to help
Men, people without mammograms or genetic data, and racial/ethnic groups not well represented in the cohorts are unlikely to benefit directly from these results.
What this research studies
From a patient's point of view, researchers are running AI algorithms on routine mammogram images and combining those image-based risk scores with genetic risk scores to create a fuller picture of breast cancer risk. They are doing this using data from two large U.S. cohorts that include many Black women so the tools work well across groups. The team will compare how well the AI-based models perform versus traditional risk tools and will test whether adding genetic information helps. The goal is to validate these approaches in real-world screening populations rather than only in specialized clinical samples.
Why it matters
Potential benefit: If successful, this work could give women more accurate and equitable personalized risk estimates to guide screening and prevention decisions.
How similar studies have gone: Early studies suggest AI on mammograms can outperform traditional risk models, but combining imaging with genetic risk scores and validating performance in large Black cohorts is largely new.
Where this research is happening
-
Columbia University Health Sciences
New York, United States Active
Researchers
- Principal investigator: Tehranifar, Parisa — Columbia University Health Sciences
- Study coordinator: Tehranifar, Parisa
Related research
Other NIH-funded projects on the same topics.
-
AKT1 mutation detection in breast cancer with AI on standard pathology slides
A new AI tool called AKT Predict aims to quickly and cheaply identify AKT1 mutations in breast cancer using regular pathology slides.
-
Using advanced imaging technology to improve breast cancer surgery outcomes
This study is testing a new imaging technology that helps doctors check the edges of tissue removed during breast surgery in real-time, aiming to reduce the chances of needing extra surgeries and improve outcomes for patients.
-
Improving breast cancer risk prediction using deep learning from mammograms
This study is working to improve how we predict breast cancer risk by using advanced computer technology on mammogram images, especially to help Black women, and it aims to create better prevention strategies by combining this information with genetic data from diverse groups of people.
-
Improving Breast Cancer Guideline Adherence in Nigeria
Aimed at helping doctors in Nigeria follow international guideline-based breast cancer care to improve outcomes for women with breast cancer.
-
Integrative histo-genomic analysis to guide cancer risk and targeted therapies with interpretable AI
Researchers are developing an advanced AI-powered approach to map genetic and cellular patterns across breast and cervical cancers to help tailor treatments.
-
Genomic study to understand how genetics and epigenetics shape triple-negative breast cancer in African ancestry
This study looks at how inherited genes and gene activity affect triple-negative breast cancer in people of African ancestry to better understand why outcomes can be worse.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.