News & Events
News
20 December 2024
Artificial Intelligence Improves Breast Cancer Diagnosis and Reduces Errors
We show that the system could be effectively integrated into the pathologic workflow as a quality assurance tool aimed to mitigate false-negative classification
New AI Tool Enhances Accuracy and Efficiency in Identifying Breast Cancer Markers.
A new study, published by a research team from the Technion, introduces an artificial intelligence (AI) system that can predict key breast cancer markers from routine tissue slides, improving diagnostic accuracy and reducing errors. This innovation was led by RTICC-affiliated member, Prof. Ron Kimmel from the Faculty of Electrical and Computer Engineering, and was published in Communications Medicine. The research, titled "Clinical Utility of Receptor Status Prediction in Breast Cancer and Misdiagnosis Identification Using Deep Learning on Hematoxylin and Eosin-Stained Slides," demonstrates how AI can assist in identifying patients eligible for hormone therapy and catching misdiagnosed cases.
A Faster, More Reliable Way to Identify Cancer Markers
Doctors rely on detecting three important proteins—estrogen receptor (ER), progesterone receptor (PR), and ERBB2 (Her2)—to determine the best treatment for breast cancer patients. Traditionally, this requires specialized tests that can be costly, time-consuming, and sometimes lead to inconsistent results. This study introduces an AI-powered system trained on nearly 20,000 tissue slides from multiple medical centers. The system can analyze routine slides stained with hematoxylin and eosin (H&E) and accurately predict the presence of these key proteins.
Catching Misdiagnoses and Improving Patient Care
One of the most significant findings of the study is the AI tool’s ability to detect cases where the original diagnosis was incorrect. By re-examining flagged samples, researchers identified 31 cases where ER, PR, or Her2 status had been misdiagnosed. This highlights the potential of AI as a second-check tool for quality assurance, helping reduce false negatives and ensuring more patients receive the right treatment. Additionally, the AI system could streamline the diagnostic process by identifying hormone receptor-positive patients without requiring additional tests.
A Step Forward for AI in Cancer Care
This study demonstrates how AI can be integrated into clinical workflows to enhance breast cancer diagnosis, reduce errors, and improve treatment planning. As artificial intelligence continues to evolve, tools like this could become an essential part of cancer care, ensuring that more patients receive accurate and timely diagnosis.
The Research has been published in the Journal Communications Medicine.
