Innovative AI model offers more effective breast cancer treatment plans

This news has been read 937 times!

NEW YORK, Nov 29: In a groundbreaking development, expert pathologists have harnessed the power of artificial intelligence (AI) to enhance breast cancer diagnosis, potentially sparing patients from unnecessary and often burdensome treatments. The innovative tool, developed by researchers at Chicago’s Northwestern Medicine in the United States, could revolutionize the approach to breast cancer treatment by providing more accurate risk estimates.

The AI tool has proven effective in identifying breast cancer patients currently categorized as high or intermediate risk who may, in fact, become long-term survivors. By offering a more precise outlook on the future course of the disease, doctors may be able to tailor chemotherapy treatments to individual patients, potentially reducing their duration or intensity. This is especially significant considering the adverse side effects, such as nausea, associated with traditional chemotherapy.

The study, published in the journal Nature Medicine, marks a milestone as the first to utilize AI for the comprehensive evaluation of both cancerous and non-cancerous elements in invasive breast cancer. While current pathology assessments primarily focus on cancerous cells to determine treatment plans, the study emphasizes the importance of analyzing patterns of non-cancerous cells in predicting outcomes.

Lee Cooper, the corresponding study author and associate professor of pathology at Northwestern University Feinberg School of Medicine, highlighted the groundbreaking nature of their findings. “Our study demonstrates the importance of non-cancer components in determining a patient’s outcome,” said Cooper. “The importance of these elements was known from biological studies, but this knowledge has not been effectively translated to clinical use.”

Female breast cancer is the most commonly diagnosed cancer in Europe, affecting more than 355,000 women in the EU in 2020 alone. With the AI tool providing higher precision risk estimates, there is potential for fewer women to undergo chemotherapy unnecessarily.

Traditionally, pathologists examine cancerous tissue to determine treatment plans based on the tissue’s appearance. However, recent studies have shown that non-cancerous cells can influence cancer cell growth. Cooper and colleagues developed an AI model that evaluates breast cancer tissue from digital images, measuring the appearance of both cancerous and non-cancerous cells and their interactions.

“The AI model measures these patterns and presents information to the pathologist in a way that makes the AI decision-making process clear to the pathologist,” explained Cooper. The tool analyzes 26 properties of a patient’s breast tissue to generate a prognostic score, providing individual scores for cancer, immune, and stromal cells to explain the overall score.

To train the AI model, researchers utilized hundreds of thousands of human-generated annotations of cells and tissue structures within digital images of tissue samples collected over several years. This innovative approach marks a significant leap forward in the quest for personalized and more effective breast cancer treatment plans.

This news has been read 937 times!