AI achieves 99% accuracy in finding metastatic breast cancer – Google

Kyle Chua
AI achieves 99% accuracy in finding metastatic breast cancer – Google
Called LYNA, the AI is able to identify notoriously hard-to-detect metastatic tumors, say Google researchers

MANILA, Philippines – Google last week shared in a blog post that it has developed an AI that could greatly aid pathologists in the diagnosis of metastatic breast cancer.

Metastatic tumors are cancerous cells that break away from their tissue of origin to travel and form new tumors in other parts of the body. They are notoriously hard to detect.

Google’s researchers hopes to address this problem with the introduction of their deep learning tool called Lymph Node Assistant, or LYNA who examines images of a patient’s cells. It was trained to recognize characteristics of tumors with the help of pathological slides that were used as datasets. Deep learning makes use of large datasets from which the AI learns. From the data it has analzyed and learned from, the AI is then able to analyze new data and diagnose the tumors. 

In tests, it reportedly managed to correctly tell apart a slide with cancer from a slide without cancer 99% of the time. The AI was also able to pinpoint the location of cancers, some of which were too small to be visible for humans.

LYNA, meanwhile, is reportedly even more effective when it’s working together with a pathologist.

Pathologists were able to more accurately detect smaller manifestations of metastases called micrometastases and reduce the rate of missed micrometastases by a factor of two with the help of the AI. LYNA’s aid can also free up time and energy for pathologists to tackle more challenging clinical tasks.

“Pathologists with LYNA assistance were more accurate than either unassisted pathologists or the LYNA algorithm itself, suggesting that people and algorithms can work together effectively to perform better than either alone,” the researchers wrote.

Despite all it can already do, the researchers admit the AI still needs some work as it has yet to be used in a real-life clinical environment. However, they remain optimistic and hope LYNA can help improve the accuracy and availability of pathologic diagnosis around the world. –

Add a comment

Sort by

There are no comments yet. Add your comment to start the conversation.