New AI-based tool improves the detection of breast cancer tumors

Researchers at Karolinska Institutet in Sweden have created an AI-based system that can improve the detection of breast cancer tumors and the ability to predict the risk of the recurrence. The higher precision in diagnosis could lead to more personalized treatment for the large group of breast cancer patients with intermediate risk tumors. The findings were published in the scientific journal Annals of Oncology.

Every year, about two million women worldwide suffer from breast cancer. The diagnostic procedure involves the tumor’s tissue samples are examined and classified by a pathologist, and classified by the risk of being low (grade 1) medium (grade 2) or high (grade 3). This helps the doctor determine the most suitable treatment for the patient.

Roughly half of breast cancer patients have an advanced tumor and, unfortunately, there is no specific guidelines regarding how the patient is to be treated. Therefore, some patients are over-treated with chemotherapy while others risk being under-treated. This is the issue we’ve attempted to solve.”

Yinxi Wang, Study First Author and Doctoral Student, Department of Medical Epidemiology and Biostatistics, Karolinska Institute

Hospitals have started to employ molecular diagnostics in a few instances to improve breast cancer risk assessment precision. However, these techniques are costly and time-consuming. The researchers at Karolinska Institutet have now developed and evaluated an AI (artificial intelligence)-based method for tissue analysis. The study shows that the AI-based method could further divide the patients with grade 2 tumors into two groups, one with high-risk and the other with low-risk that can be clearly distinguished in terms of the risk of recurrence.

“One major benefit of this method is that it’s cost-effective and fast, since it’s based on microscope images of dyed tissue samples, which are already a an integral part of hospital procedures,” says co-last author Johan Hartman, professor of pathology at the Department of Oncology-Pathology, Karolinska Institutet and a pathologist at the Karolinska University Hospital. “It allows us to provide this type of diagnosis to more people and improves our ability to provide the best treatment to every patient.”

The AI model was trained to recognize the characteristics of high-resolution microscopic pictures of patients suffering from grade 1 and 3 tumors. The study is based upon a large collection of microscopic images that includes 2,800 tumors.

“It’s amazing that deep learning can develop models that not just replicate what specialist doctors do, but also allow them to extract information beyond the human eye,” said Mattias Rantalainen (co-last author), associate professor and director of the research group at the Department of Medical Epidemiology and Biostatistics of Karolinska Institutet.

While the method isn’t yet ready to be utilized in clinical trials, a product that has been approved by regulators is being developed by Stratipath AB. This company is supported by KI Innovations. The method will be further evaluated by the researchers with the goal of bringing the product to market in 2022.

Journal reference:

Wang, Y. and. al. (2021) Improved the histological classification of breast cancer using deep learning. Annals of Oncology.

Content Source:

Gemma Wilson

Gemma is a journalism graduate with keen interest in covering business news – specifically startups. She has as a keen eye for technologies and has predicted quite a few successful startups over the last couple of years.

Related Articles