IBM's AI distinguishes between healthy, cancerous cells in breast tumors


Using mass cytometry and machine learning technology, scientists at IBM's Zurich-based research lab and the University of Zurich have mapped out the cellular composition of breast cancer tumors, according to a study published in Cell this week.

Researchers studied breast tumor and non-tumor samples from 144 patients, utilizing single-cell mass cytometry to quantify the 70 proteins in each of 26 million individual cells before turning to machine learning processes to seek out patterns and relationships in that data. With those models, the scientists were able to create a "detailed atlas" of breast cancer ecosystems.

This analysis and resulting atlas prove that each breast cancer tumor carries a unique cellular composition, highlighting the need for individualized, precision treatments. The research also provides insight into those potential treatments: The discovery of certain immune cell types in one group of the studied tumors, for example, suggests that immunotherapy could be an effective treatment for some breast cancer patients.

The study's authors wrote in a blog post that they plan to initiate subsequent studies into the proposed immunotherapy treatments, with the hope of achieving results that could eventually lead to clinical trials.

More articles about AI:
UPenn Medicine's AI tool for data analytics is open-source, free to the public
AI hearing aid uses brain activity to amplify specific voices
AI can predict patient response to popular Crohn's treatment

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Whitepapers

Featured Webinars