skip to main content
Research Highlight | October 04, 2019

Artificial intelligence helps researchers predict if breast cancer patients will respond to chemotherapy before treatment is administered

A team of TFRI-funded researchers based out of the Sunnybrook Research Institute in Toronto has devised a new technique that could determine if patients diagnosed with locally advanced breast cancer will respond to neoadjuvant chemotherapy (NAC), days before the treatment is even administered.

The new technique uses a form of artificial intelligence known as an artificial neural network (ANN) to analyze images of tumours collected with quantitative ultrasound. To get a result, the team “fed” the ANN a series of ultrasound images as well as corresponding patient outcome data, and then “taught” it to find patterns between the images and the outcomes.

After this, the ANN could analyze an ultrasound image prior to treatment and predict whether the patient would respond to chemotherapy with 96 per cent (+/- 6 per cent) accuracy, according to the study published in Oncotarget (June 2019).

 “Our findings provide a framework for developing personalized a priori chemotherapy selection for patients that are candidates for NAC, potentially resulting in improved patient treatment outcomes and prognosis,” said Dr. Gregory Czarnota, a radiation oncologist at Sunnybrook and leader of a Terry Fox New Frontiers Program Project Grants.

According to Dr. Czarnota, knowing if a patient with locally advanced breast cancer will respond positively to chemotherapy before the treatment is administered can help avoid months of unnecessary treatments that are often accompanied with negative side effects. It can also force doctors to find alternative treatments before the disease progresses, which can improve outcomes and prognosis.

“This finding not only opens the door for more personalized treatments for cancer patients with breast cancer but also serves as a proof-of-concept of the role artificial intelligence can play in improving cancer diagnosis and treatment,” said Dr. Czarnota.


A priori prediction of breast tumour response to chemotherapy using quantitative ultrasound imaging and artificial neural networks


Hadi Tadayyon, Mehrdad Gangeh, Lakshmanan Sannachi, Maureen Trudeau, Kathleen Pritchard, Sonal Ghandi, Andrea Eisen, Nicole Look-Hong, Claire Holloway, Frances Wright, Eileen Rakovitch, Danny Vesprini, William Tyler Tran, Belinda Curpen and Gregory Czarnota


This study was partially find by The Terry Fox New Frontiers Program Project Grant in Ultrasound and MRI for Cancer Therapy