Khaberni - A research team has unveiled an artificial intelligence model capable of predicting the extent to which breast cancer patients benefit from chemotherapy, based on traditional tissue images.
According to a study published in The Lancet Oncology, this approach could help in reducing unnecessary treatment and speeding up medical decision-making.
Determining the need for chemotherapy after surgery is one of the toughest challenges, as the majority of patients do not benefit from it despite its side effects. Here, the new model plays a role, analyzing tissue biopsy images instead of relying on costly genetic tests.
The researchers relied on data from more than 10,000 patients in a large random clinical trial, allowing for an accurate assessment of the model's ability to predict not only the risk of disease recurrence but also the actual benefit from the treatment.
The system uses deep learning techniques to analyze high-resolution images of the tumor tissues. It evaluates complex patterns such as cell division, tissue structure, and immune response, which are signals difficult to accurately measure with the human eye.
Within minutes, the model produces a numerical score that helps doctors and patients make a joint decision regarding treatment.
Accuracy, speed, and lower cost
Results show that the model works efficiently across different health systems, after being tested on thousands of cases in several countries. It is characterized by swift execution, as it only requires digital scanning of already available samples, without additional procedures or long waiting times.
The researchers believe this technology could be crucial in resource-limited countries where genetic testing is scarce.
This model represents a step towards more personalized medicine, as it helps identify patients who really need chemotherapy, potentially improving outcomes and reducing unnecessary therapeutic burdens.
