Harmonizing MR Images to Improve Robust Federated Learning in MS
Official title: Retrospective Magnetic Resonance Image Harmonization for Robust Federated Learning
Funded at Henry M. Jackson Fdn for the Adv Mil/med · NIH-11378675 · Data updated 2026-11-12
A research project will test methods that make brain MRI analysis more accurate across different clinics by using privacy-friendly learning together with image harmonization, focusing on people with multiple sclerosis.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Henry M. Jackson Fdn for the Adv Mil/med NIH-funded |
| Lab location | 1 site (Bethesda, United States) |
| Project start | July 28, 2025 |
| Project end | May 31, 2029 |
| Funding year | 2026 |
| Award (this year) | $326,043 |
| NIH project number | 5R01LM014784-02 |
| Project ID | NIH-11378675 on NIH RePORTER |
Who could benefit from this research
✓Could help
Ideal candidates are adults with multiple sclerosis whose MRI data could be included in collaborative, privacy-preserving analyses.
✕Less likely to help
Patients whose MRI protocols are not compatible with the harmonization methods may see limited benefit.
What this research studies
This project aims to make deep learning models for brain MRI segmentation more reliable when images come from different machines and settings. By combining federated learning with image harmonization, the team will reduce differences caused by hardware and software and improve lesion quantification in MS patients. The study uses retrospective data from MS patients to develop and test methods that work well across diverse MRI datasets while protecting patient privacy. The work is designed to translate into more consistent imaging biomarkers that clinicians can trust in real-world settings.
Why it matters
Potential benefit: If successful, this approach could lead to more accurate MRI-based measurements of MS disease activity across clinics, supporting better treatment decisions.
How similar studies have gone: Previous work shows promise for domain adaptation and federated learning in medical imaging, but robust cross-site performance remains challenging.
Where this research is happening
-
Henry M. Jackson Fdn for the Adv Mil/med
Bethesda, United States Active
Researchers
- Principal investigator: Pham, Dzung L — Henry M. Jackson Fdn for the Adv Mil/med
- Study coordinator: Pham, Dzung L
Related research
Other NIH-funded projects on the same topics.
-
Cooperation partners of SETBP1 in AML development
This grant explores how SETBP1 and AF9 (MLLT3) work together to drive acute myeloid leukemia and whether targeting related epigenetic regulators could help treat aggressive CK-AML.
-
U.S. Military HIV Research Program Clinical Trials Unit expansion
A nationwide network of clinical trial sites is being expanded to test HIV prevention and treatment approaches through DAIDS-supported trials.
-
Novel protein biomarkers of chronic pain in adults using a pain registry biobank
A study in adults with chronic pain will look for new protein biomarkers in blood samples to understand better why pain persists and how to improve treatment.
-
Tailored antibody cures for HIV using AI and vaccines
A study to design personalized antibody-based strategies to achieve HIV remission, using lab experiments and computer models to tailor vaccines or antibody treatments to individuals or small groups.
-
Understanding how antibody protection works using 3D protein structures
Researchers will build a 3D-based method to see which viral features relate to prevention by antibodies, using data from people in HIV vaccine and monoclonal antibody trials.
-
Military HIV Clinical Network
This program runs HIV and TB prevention and treatment clinical studies for adults in military-associated and high-HIV-burden communities.
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.