Oguzhan Alagoz, the Proctor and Gamble Bascom Professor in the Department of Industrial and Systems Engineering at the University of Wisconsin-Madison, has received a new $2.4 million grant from the National Cancer Institute to apply mathematical modeling to the overdiagnosis of thyroid cancer.
Advances in imaging technology in recent decades have revealed more small cancerous thyroid nodules in patients, findings that can prompt biopsies, removals and other procedures. But for otherwise asymptomatic patients with low-risk cancerous nodules, those treatments may not be necessary.
“It turns out that 90-95% of these surgeries could be unnecessary,” says Alagoz.
During the nearly five-year grant, Alagoz will work with David Francis, an associate professor of surgery, to first build a simulation model to estimate the scope of overdiagnosis. Alagoz says it will be the first time mathematical modeling has been applied to quantify this particular issue.
Then, Alagoz and Francis will use a novel machine learning method to inform a model they’ll develop to test alternative treatment approaches, such as regular monitoring of nodules below a certain size threshold rather than surgical removal.
“Once we have a mathematical model that represents how fast these tumors grow and which ones are going to be fatal and which ones are completely harmless, then we can actually reduce the number of surgeries and we can spare a lot of people from unnecessary surgeries and a lot of costs,” says Alagoz.
Author: Tom Ziemer