AI to help spot aggressive versus slow prostate cancer on MRI
Official title: Rad-pathomic deep learning models to assist radiologists in differentiating aggressive from indolent prostate cancer on MRI
Funded at Stanford University · NIH-11252779 · Data updated 2026-11-12
This project uses artificial intelligence that links MRI images and pathology to help doctors tell aggressive prostate cancer from slow-growing or benign tissue for men facing biopsy decisions.
Quick facts
| Grant type | R37 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Stanford University NIH-funded |
| Lab location | 1 site (Stanford, United States) |
| Project start | January 12, 2022 |
| Project end | December 31, 2026 |
| Funding year | 2026 |
| Award (this year) | $533,076 |
| NIH project number | 5R37CA260346-05 |
| Project ID | NIH-11252779 on NIH RePORTER |
Who could benefit from this research
✓Could help
Men undergoing prostate MRI because of elevated PSA, abnormal exam, or who are considering a prostate biopsy would be the ideal candidates.
✕Less likely to help
Men without a usable prostate MRI, those with MRI-incompatible implants, or people whose cancer is already fully treated and under stable management may not benefit directly.
What this research studies
This project develops an AI that links features seen on your prostate MRI to what pathologists see under the microscope, using matched MRI and biopsy images. The team will train deep-learning models to create 'pathomic' MRI biomarkers that highlight areas more likely to be aggressive cancer. They will test the models' ability to detect cancer and to tell aggressive from indolent lesions, including when both types are present in the same lesion. The aim is to produce tools that can flag suspicious areas on MRI to guide biopsy and reduce unnecessary procedures.
Why it matters
Potential benefit: If successful, this could help find life-threatening prostate cancers earlier while reducing unnecessary biopsies and their side effects.
How similar studies have gone: Early work by this team and other MRI–AI studies have shown promising accuracy (AUCs around 0.85), but combining pathology-derived biomarkers with MRI to localize aggressive versus indolent disease is relatively new.
Where this research is happening
-
Stanford University
Stanford, United States Active
Researchers
- Principal investigator: Rusu, Mirabela — Stanford University
- Study coordinator: Rusu, Mirabela
Related research
Other NIH-funded projects on the same topics.
-
Deep learning to improve MRI-based prostate cancer detection
A new AI tool helps doctors interpret prostate MRI to more accurately find clinically significant cancer in American men.
-
Deep learning MRI biomarkers to predict prostate cancer outcomes
A Yale-led project uses advanced MRI images and artificial intelligence to better predict how localized prostate cancer will behave in patients.
-
Using deep learning to help identify aggressive prostate cancer on MRI
This study is working on making MRI scans better at telling the difference between aggressive and non-aggressive prostate cancer, which could help reduce unnecessary biopsies and improve how doctors detect cancer, ultimately leading to better treatment choices for patients.
-
Using advanced digital imaging and AI to improve prostate cancer diagnosis
This study is working on using advanced computer technology to help doctors better understand prostate cancer by analyzing tissue samples, making it easier to spot tumors and explain the results clearly.
-
Using advanced imaging and AI to improve prostate cancer biopsy decisions
This study is working on new tools that use advanced imaging and artificial intelligence to help doctors better diagnose prostate cancer, making it easier to decide who needs a biopsy and where to focus on in the prostate.
-
Cardiovascular risk planning and reduction for men with prostate cancer
A program to better predict and reduce heart disease risk in men treated for prostate cancer, especially older patients.
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.