AI algorithm may detect pancreatic cancer as accurately as radiologists
Elliot Fishman, MD, a professor of radiology at Johns Hopkins Medicine in Baltimore, is hoping to develop an artificial intelligence (AI) program to detect pancreatic tumors earlier, according to an article published May 30 by NPR. He is working with computer science students to design a tumor-detecting AI algorithm that can be built into computed tomography (CT) scanner software to recognize and differentiate between a normal pancreas and one that’s cancerous.
"That's the ultimate opportunity—to be able to diagnose it before you have any symptoms and at a stage where it's even maybe too subtle for a radiologist to be able to detect it," said Karen Horton, MD, chair of the Johns Hopkins radiology department and collaborator on the project.
The project, dubbed the Felix Project, has been in the making for over a year as researchers continue to feed their annotated CT scans into the project's computer to teach it to recognize the signs of a tumor that experienced radiologists can identify in the scans.
To date, project researchers have fed the computer almost 2,000 scans of both healthy people and those with pancreatic cancer, according to NPR.
Read more below:
A recent graduate from Dominican University (IL) with a bachelor’s in journalism, Melissa joined TriMed’s Chicago team in 2017 covering all aspects of health imaging. She’s a fan of singing and playing guitar, elephants, a good cup of tea, and her golden retriever Cooper.