UK-based oncologists have developed an AI model that helps predict whether aggressive types of breast cancer will spread based on changes in a patient’s lymph nodes.
The study was published on Thursday within the Journal of Pathology by Breast Cancer Now and funded by researchers at King’s College of London.
Secondary or “metastatic breast cancer” refers to when breast cancer cells have spread to other parts of the body.
While it will possibly be treated, it can’t be cured.
The scientists behind last week’s study hope that by utilizing artificial intelligence to investigate the immune response within the lymph nodes of ladies with triple-negative breast cancer, they will find a way to raised assess the likelihood of the disease spreading.
Lymph nodes are lumps of tissue throughout the body that are critical in helping the body fight infection and disease.
If breast cancer cells begin to spread, patients often need more intensive treatment.
Still, scientists imagine it is feasible to predict whether cancer cells will spread based on their immune response.
![mammography](https://nypost.com/wp-content/uploads/sites/2/2023/05/NYPICHPDPICT000010821043.jpg?w=1024)
Dr. Anita Grigoriadis, who led the research, said the team took the outcomes under the microscope and “translated them right into a deep learning framework to create an AI model that has the potential to help doctors treat and take care of patients by providing them with one other tool of their arsenal to help prevent secondary breast cancer.”
Researchers tested the AI model on over 5,000 lymph nodes donated by nearly 350 patients to biobanks.
“By showing that lymph node changes can predict the spread of triple-negative breast cancer, we have now leveraged our growing understanding of the necessary role that the immune response can play in understanding a patient’s prognosis,” said Grigoriadis.
![Artificial intelligence helps pathologists to determine the overall percentage of tumors](https://nypost.com/wp-content/uploads/sites/2/2023/05/NYPICHPDPICT000011898195.jpg?w=1024)
The team plans to further test the model in centers across Europe.
“Moving from evaluating tissue on slides under a microscope to using computers [National Health Service] is gaining momentum,” said Grigoriadis. “We would like to make use of this modification to develop AI-driven software based on our model for pathologists that can profit women with this difficult-to-treat breast cancer.”