AI-based tool set to improve breast cancer diagnosis

1 min read

An AI-based tool that improves breast cancer diagnosis and predicts the risk of recurrence has been developed by researchers in Sweden.

AI-based tool
Image by Colin Behrens from Pixabay

The advance from a team at the Karolinska Institutet could lead to more personalised treatment for breast cancer patients with intermediate risk tumours. The results are published in Annals of Oncology.

In the diagnostic procedure for breast cancer, tissue samples of the tumour are analysed and graded by a pathologist and categorised by risk as low (grade 1), medium (grade 2) or high (grade 3), which guides decisions on the most suitable treatment.

“Roughly half of breast cancer patients have a grade 2 tumour, which unfortunately gives no clear guidance on how the patient is to be treated,” said Yinxi Wang, a doctoral student at the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet. “Consequently, some of the patients are over-treated with chemotherapy while others risk being under-treated. It’s this problem that we’ve tried to resolve.”

The AI-based tool further divides the patients with grade 2 tumours into two sub-groups - one high-risk and one low-risk - that are clearly distinguishable in terms of the recurrence risk.

“One big advantage of the method is that it’s cost-effective and fast, since it’s based on microscope images of dyed tissue samples, which is already part of hospital procedure,” said Johan Hartman, professor of pathology at the Department of Oncology-Pathology, Karolinska Institutet, and pathologist at the Karolinska University Hospital. “It enables us to offer this type of diagnosis to more people and improves our ability to give the right treatment to any one patient.”

The AI model has been trained to recognise characteristics of high-resolution microscopic images from patients classified with grade 1 and grade 3 tumours. The study is based on an extensive microscopic image bank of 2,800 tumours.

“It’s fantastic that deep learning can help us develop models that don’t just reproduce what specialist doctors do today, but also enable us to extract information beyond the scope of the human eye,” said Mattias Rantalainen, associate professor and research group leader at the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet.

The method is not ready for clinical application, but a regulatorily approved product is under development by Stockholm-based Stratipath. The researchers will now be further evaluating the method with the aim to have a product out on the market by 2022.