AI helps determine if patient’s breast cancer could return after treatment, study shows

CLEVELAND (WJW) – Breast cancer is the second most common cause of cancer deaths in women in the United States, with 80% of patients diagnosed in the early stages.

While most cases are treated aggressively through either chemo or radiation, what if doctors can predict if cancer will return?

“What we’ve identified is that there are features, patterns in the biopsies, that can actually tell us as to which cancers are more aggressive,” said Anant Madabhushi, a professor of biomedical engineering at Case Western Reserve University.

Madabhushi is the lead scientist in this two-year groundbreaking study, using artificial intelligence to predict whether cancer will come back.

Madabhushi says patients who had less aggressive cancer tended to have a more chaotic arrangement of collagen, a common protein found throughout the body.

“Women who tended to have more aggressive breast cancer with the worst outcome were the women who had more structured, ordered collagen,” Madabhushi said.

Madabhushi added that 60% of newly diagnosed cancer patients in the United States will go bankrupt for treatments that may not even be necessary.

“This strategy could potentially alleviate toxicity because of the chemotherapy, but also the financial toxicity,” Madabhushi said.

It could potentially bring a bright future where breast cancer patients will no longer worry if or when cancer could return.

Tissue samples from more than 700 patients in Cleveland and worldwide were used as part of this study.

Doctors are now examining if using certain drugs could help destruct collagen in the body.

Source: Fox 8 by: Jennifer Jordan