Counterfactual explanations for AI-assisted cancer diagnosis and subtyping
Official title: SCH: Counterfactual Explanations for AI-Assisted Cancer Diagnosis and Subtyping
Funded at University of Illinois at Urbana-Champaign · NIH-11637021 · Data updated 2026-11-12
A cancer-diagnosis AI tool will be made easier to understand by showing counterfactual explanations that reveal how changing histopathology images could change the model’s prediction.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Illinois at Urbana-Champaign NIH-funded |
| Lab location | 1 site (Champaign, United States) |
| Project start | July 1, 2024 |
| Project end | June 30, 2028 |
| Funding year | 2026 |
| Award (this year) | $282,425 |
| NIH project number | 7R01CA297832-03 |
| Project ID | NIH-11637021 on NIH RePORTER |
Who could benefit from this research
✓Could help
Not applicable for this grant, as it focuses on method development and explanations for AI models using pathology data.
✕Less likely to help
No direct patient participation is required for this methodological research.
What this research studies
This project focuses on making AI-based cancer diagnosis more transparent for pathologists by developing counterfactual explanations for deep learning models that analyze whole-slide histopathology images. The work aims to create a principled framework that can explain why the model classifies tumors as non-aggressive or aggressive, not just where it looks. Researchers will build a dataset-level explainer and test how these explanations influence clinical interpretation and trust in AI-assisted workflows. The study is designed to improve the usability and adoption of AI tools in pathology to support more accurate and timely treatment decisions.
Why it matters
Potential benefit: If successful, the explanations could help pathologists understand AI recommendations better, potentially leading to more accurate diagnoses and treatment planning.
How similar studies have gone: Previous work has shown that explainable AI can influence clinician trust, though applying counterfactual explanations to whole-slide pathology is novel.
Where this research is happening
-
University of Illinois at Urbana-Champaign
Champaign, United States Active
Researchers
- Principal investigator: Wang, Hao — University of Illinois at Urbana-Champaign
- Study coordinator: Wang, Hao
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.