Lead researchers
This research was led by:
Why did we do this research?
Most colorectal cancers develop from smaller growths called colorectal polyps, and the best way to fight colorectal cancer is by removing colorectal polyps during colonoscopy before they progress to cancer. Polyps often reoccur, and the time between colonoscopies depends on the polyp type diagnosed during microscopic examination by pathologists. Some polyp types may progress to colorectal cancer faster than other types. Unfortunately, deciding what kind of polyp has been removed can be challenging, and there is a degree of variability among pathologists on how to classify polyps.
What did the research involve?
Drs. Hassanpour and Suriawinata built an artificial intelligence (AI) model to automatically identify and classify polyp types on high-resolution images to help clinicians standardize polyp diagnosis and develop individualized follow-up plans.
What did we learn?
The AI model can quickly and accurately classify different types of polyps on high-resolution images. The model displays the polyp clearly and classifies polyps as accurately as expert pathologists.
Why is this important?
The AI model can help pathologists make a more accurate and standardized classification of different polyp types removed during colonoscopy and improve plans for patient follow-up plans in the fight against colorectal cancer.
To learn more about this research
Please refer to this news release on the EurekAlert! website to learn more about this research.
Funding acknowledgement
This research was made possible with funding from the National Institutes of Health.
Special thanks
We want to thank our Community Research Ambassadors Tawny and Terry for partnering with us to develop the video and content for this page. Thanks, Tawnya and Terry!