University of Minnesota researchers hope to limit heart issues related to breast cancer treatment using AI

Researchers look to AI to help breast cancer patients

MINNEAPOLIS — Researchers at the University of Minnesota are trying to limit heart issues that are sometimes a side effect of treating breast cancer with a little help from AI.

If a doctor determines someone has breast cancer, the next step is treatment — but certain kinds, like chemotherapy, can cause heart damage in some patients. 

Rui Zhang, researcher at the University of Minnesota Medical School, says there's a gap in the data to explain who is most at risk for this side effect and why.

"Currently, there's limited study focused on the cardiovascular disease prediction within the cancer treatments, the patients," Zhang said.

RELATED: How AI is bringing new options to mammograms, other breast cancer screenings

Zhang and Assistant Computer Science Professor Ju Sun are working on leveraging artificial intelligence to help doctors more accurately predict heart issues in breast cancer patients.

"Complement, inform and augment doctors. That's what I think AI for health care can do," Sun said.

But the AI tool to accomplish their goals doesn't exist yet.

Sun says AI right now works best when there's tons of information it can learn from.

Their work at the U is focused on "rewiring" an AI model to work with more limited and variable data about heart problems linked to cancer treatment, since every patient is different.

"So whenever things are underrepresented, AI tends to not work well," Sun said. "So that's a technical problem we're trying to tackle, you know, in this project, so basically, to ensure, you know, AI is performing in a fair way."

The National Institute of Health awarded the U of M researchers $1.2 million in grant funding to complete the project over the next four years.

The work is ongoing, but if they're successful, the AI model could be used for other cancers and diseases too.

"So I think that's also one of the very innovation and then potentially this could be implemented and actually improve the healthcare outcomes," Zhang said.

Read more
f

We and our partners use cookies to understand how you use our site, improve your experience and serve you personalized content and advertising. Read about how we use cookies in our cookie policy and how you can control them by clicking Manage Settings. By continuing to use this site, you accept these cookies.