Reducing bias in AI used for breast cancer screening
SCH: Quantifying and mitigating demographic biases of machine learning in real world radiology
This project will make AI tools used in breast cancer screening work more fairly and accurately for people of different ages and breast tissue types.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Johns Hopkins University NIH-funded |
| Lab location | 1 site (Baltimore, United States) |
| Project ID | NIH-11126591 on NIH RePORTER |
What this research studies
Researchers at Johns Hopkins will analyze real-world mammogram images and how current AI tools perform across different groups. They will build tools to find and measure differences in AI accuracy that come from different scanners, protocols, or tissue characteristics, even when those details are not recorded. The team will develop algorithms to correct for the worst-case gaps in performance so the AI is fairer across populations. The focus is on improving screening for breast cancer using clinical imaging data from routine care.
Who could benefit from this research
Good fit: People who have had mammograms or participate in breast cancer screening programs, especially those from varied age groups and breast tissue types, would be most relevant for the data used in this work.
Not a fit: People without mammograms, those whose care does not involve AI-based screening, or individuals with conditions unrelated to breast imaging are unlikely to see direct benefits from this project.
Why it matters
Potential benefit: If successful, this work could help AI screening tools detect breast cancer more reliably across diverse patient groups, reducing missed cancers or unnecessary follow-ups for some people.
How similar studies have performed: Previous studies have shown that AI can be biased in medical imaging and that some correction methods help, but this proposal focuses on newer techniques to fix hidden sources of variability that have been less tested in real-world screening.
Where this research is happening
Baltimore, United States
- Johns Hopkins University — Baltimore, United States (Active)
Researchers
- Principal investigator: Sulam, Jeremias — Johns Hopkins University
- Study coordinator: Sulam, Jeremias
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 below.