Adapting a machine learning tool to read thyroid biopsies across diverse settings
Official title: Adapting a machine learning algorithm to predict thyroid cytopathologyin LMIC
Funded at Duke University · NIH-11535317 · Data updated 2026-11-12
A machine learning tool will help quickly and accurately read thyroid biopsy slides from smartphone images, aiming to speed up diagnosis and reduce errors for patients in the United States and in LMIC partner sites.
Quick facts
| Grant type | NIH-funded research |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Duke University NIH-funded |
| Lab location | 1 site (Durham, United States) |
| Project start | August 1, 2021 |
| Project end | July 31, 2027 |
| Funding year | 2026 |
| Award (this year) | $12,960 |
| NIH project number | 3R33CA268428-05S1 |
| Project ID | NIH-11535317 on NIH RePORTER |
Who could benefit from this research
✓Could help
Patients with thyroid nodules undergoing ultrasound-guided FNAB who could benefit from faster, more consistent cytopathology interpretations.
✕Less likely to help
Patients at centers not implementing the tool or where the workflow changes are not feasible may not experience direct benefits.
What this research studies
This project will adapt a smartphone-based machine learning algorithm to classify thyroid FNAB slides, addressing variability in preparation and staining across centers. It will stress-test the model using data from Duke and partner centers in Vietnam and Tanzania to ensure reliability in diverse clinical settings. The work focuses on making diagnostic results faster and more consistent so patients don’t wait for care or undergo unnecessary surgery. The effort combines AI development with real-world testing in multiple populations to improve thyroid cancer care pathways. If successful, clinicians could rely on a robust tool to support timely decision-making.
Why it matters
Potential benefit: The research could shorten time-to-diagnosis and reduce misclassification of thyroid nodules, leading to better patient outcomes.
How similar studies have gone: Prior work shows promise in AI-assisted cytopathology, but this study expands validation across diverse staining methods and international sites.
Where this research is happening
-
Duke University
Durham, United States Active
Researchers
- Principal investigator: Lee, Walter Tsong — Duke University
- Study coordinator: Lee, Walter Tsong
Related research
Other NIH-funded projects on the same topics.
-
Investigating genetic factors in Lewy body diseases
This study is looking at how changes in a specific gene might play a role in conditions like Parkinson's disease and dementia, so that people can better understand how their genetics affect their health and possibly find more tailored treatments.
-
Targeting the PAX3-FOXO1 protein in a deadly childhood cancer
This study is looking at a specific protein that helps a tough childhood cancer called alveolar rhabdomyosarcoma, with the goal of finding new ways to treat kids who have this illness.
-
Improving population health by studying reproduction and development
This study is looking at how different things in our environment can impact pregnancy and the health of babies, and it’s for people who are planning to have children or are already parents, as it aims to help improve understanding and outcomes for everyone involved.
-
Modeling axon degeneration in C. elegans to help ALS patients
Researchers are using a tiny worm model to uncover how axons break down, which could point to new ways to help people with ALS and related nerve diseases.
-
Induced Suppression of Platelet Activity in Aneurysmal SAH Management
A new approach tests whether carefully reducing platelet activity can improve outcomes after aneurysmal subarachnoid hemorrhage.
-
AI-enhanced virtual reality tool to support remote emergency brain surgery training in Tanzania
A new AI-powered virtual reality tool helps non-experts in low-resource areas get real-time remote guidance for emergency traumatic brain injury surgery.
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.