AI to predict melanoma coming back using tumor images, gene data, and pathology
Official title: Explainable Artificial Intelligence for Melanoma Recurrence Prediction via Integrative Modeling of Multiplexed Imaging with Transcriptomics and Histopathology
Funded at University of Chicago · NIH-11457144 · Data updated 2026-11-12
This project is building an AI tool that uses detailed tumor pictures, gene activity, and pathology reports to help doctors predict whether early-stage (stage I or II) melanoma might return.
Quick facts
| Grant type | NIH-funded research |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Chicago NIH-funded |
| Lab location | 1 site (Chicago, United States) |
| Project start | April 1, 2026 |
| Project end | March 31, 2029 |
| Funding year | 2026 |
| Award (this year) | $249,000 |
| NIH project number | 4R00CA286966-03 |
| Project ID | NIH-11457144 on NIH RePORTER |
Who could benefit from this research
✓Could help
Ideal candidates are people diagnosed with stage I or II melanoma who have available tumor biopsy or surgical tissue and agree to have their samples and medical records used for analysis.
✕Less likely to help
Patients with advanced (stage III/IV) melanoma, those without available tumor tissue, or those unwilling to provide tissue or records are unlikely to benefit directly from this project.
What this research studies
This project will combine high-resolution tumor imaging, spatial maps of gene activity, and standard pathology slides to look for patterns linked to recurrence. Researchers will use explainable artificial intelligence so the models can show which cells, interactions, or gene signals drive their predictions. They will train and test the models using clinical records and tissue samples from people with early-stage melanoma and refine the methods to improve accuracy and clarity. If the models work well, the team plans further validation so the tools could be used in clinical settings to guide follow-up and treatment.
Why it matters
Potential benefit: If successful, this could help identify early-stage melanoma patients at higher risk of recurrence so they can get closer monitoring or earlier treatment.
How similar studies have gone: Some prior work has used AI or genomics for melanoma prognosis, but combining multiplexed imaging, spatial transcriptomics, and explainable AI for early-stage recurrence prediction is a newer approach with limited prior real-world validation.
Where this research is happening
-
University of Chicago
Chicago, United States Active
Researchers
- Principal investigator: Wan, Guihong — University of Chicago
- Study coordinator: Wan, Guihong
Related research
Other NIH-funded projects on the same topics.
-
Dormant melanoma cells and aging: how the tissue environment shapes recurrence
This NIH grant studies how aging tissue environments influence melanoma dormancy and late recurrences, with a focus on how aged lung fibroblasts and Wnt signaling affect cancer cell reawakening.
-
Methylation test to predict melanoma recurrence after surgery
A patient-facing test uses a DNA methylation pattern to help doctors predict if stage II melanoma might come back after surgery.
-
A new test to predict melanoma recurrence using DNA analysis
This study is looking to create a new test that checks DNA markers in patients with stage II melanoma to better predict if their cancer might come back, helping doctors decide who might need extra treatment and who can avoid unnecessary ones.
-
Using advanced AI to predict melanoma recurrence by analyzing imaging and genetic data.
This study is looking to help people with early-stage melanoma by using advanced imaging and genetic information to better predict if their cancer might come back, so they can receive more personalized care and monitoring based on their specific risk.
-
Keloid recurrence study using multi-omics in diverse patients
A research effort will look at why keloids come back after surgery, especially in African American patients, using comprehensive molecular data from patient samples.
-
Combined Imaging and RNA Markers to Predict Head and Neck Cancer Recurrence
A research project aims to use imaging and RNA data to better predict if head and neck cancer will come back, helping doctors tailor treatments for 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.