Visual explanation-guided learning for human-AI cancer imaging
Official title: SCH: Visual explanation-guided learning for human-AI collaborative abdominal cancer diagnostic imaging
Funded at Emory University · NIH-11397414 · Data updated 2026-11-12
A new AI-powered system helps doctors find abdominal cancers earlier by using visual explanations to guide how the AI analyzes CT/MRI scans.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Emory University NIH-funded |
| Lab location | 1 site (Atlanta, United States) |
| Project start | September 15, 2024 |
| Project end | August 31, 2028 |
| Funding year | 2026 |
| Award (this year) | $290,511 |
| NIH project number | 5R01CA297856-03 |
| Project ID | NIH-11397414 on NIH RePORTER |
Who could benefit from this research
✓Could help
Patients who are at risk for abdominal cancers or who undergo abdominal imaging for cancer screening or diagnosis.
✕Less likely to help
Patients without abdominal cancer symptoms or those not undergoing imaging may not directly benefit from this AI tool.
What this research studies
This project develops a computer-aided imaging tool that learns from large collections of abdominal cancer images and communicates its reasoning to clinicians through visual explanations. It aims to make AI-assisted interpretations more sample-efficient and easier to understand, reducing the likelihood of missed small tumors. The system focuses on improving collaboration between medical experts and AI by aligning the model’s reasoning with clinical judgment while protecting patient privacy. If successful, this approach could speed up accurate early detection and help guide decisions about further testing or treatment.
Why it matters
Potential benefit: Could lead to earlier, more reliable detection of abdominal cancers, potentially improving outcomes.
How similar studies have gone: Similar AI explainability approaches have shown promise but still face challenges; this work seeks to advance practical, clinically interpretable AI for cancer imaging.
Where this research is happening
-
Emory University
Atlanta, United States Active
Researchers
- Principal investigator: Zhao, Liang — Emory University
- Study coordinator: Zhao, Liang
Related research
Other NIH-funded projects on the same topics.
-
Enabling Imaging and Alpha Therapy with New Chelators for Cancer
A cancer-focused project aims to create new chelators that let alpha-emitter radiopharmaceuticals both image and treat tumors more effectively.
-
AI-powered same-day read to speed breast cancer screening results
A new AI system is being tested to give immediate screening results after a mammogram to reduce recalls, shorten anxiety, and speed up follow-up for breast cancer screening.
-
Bedside retinal imaging to detect and track retinopathy of prematurity
This project aims to bring easy-to-use bedside retinal imaging for premature infants to help find and track the blood vessel changes that can lead to childhood blindness.
-
AI for same-day breast screening results
This project will use artificial intelligence to give people having screening mammograms faster same-day results and reduce unnecessary callbacks.
-
Explainable AI to help doctors detect abdominal cancers on scans
This project trains AI to highlight and explain suspicious areas on CT and MRI scans to help doctors find abdominal cancers earlier.
-
Using AI to improve same-day breast cancer imaging results
This study is looking at how using artificial intelligence can make breast cancer screenings faster and more accurate, helping to reduce the stress of false alarms for patients and ensuring that those who need more tests get them quickly.
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.