STAT

Google’s AI improves accuracy of lung cancer diagnosis, study shows

In early testing, Google's #artificialintelligence system outperformed six radiologists in diagnosing lung cancer from reviewing a single CT scan.

One of lung cancer’s most lethal attributes is its ability to trick radiologists. Some nodules appear threatening but turn out to be false positives. Others escape notice entirely, and then spiral without symptoms into metastatic disease.

On Monday, however, Google unveiled an artificial intelligence system that — in early testing — demonstrated a remarkable talent for seeing through lung cancer’s disguises.

A published in Nature Medicine reported that the algorithm, trained on 42,000 patient CT scans taken during a National Institutes of Health clinical trial, outperformed six radiologists in determining whether patients had cancer. It detected 5% more cancers and cut false positives —

You’re reading a preview, subscribe to read more.

More from STAT

STAT2 min read
STAT+: Pharmalittle: We’re Reading About A 486% Drug Price Hike, AstraZeneca CEO Pay, And More
Is there any circumstance when a 486% increase for a medicine might be justified? A small company called Harrow argues the case for an injectable eye treatment.
STAT1 min read
Disparities In Donor Acceptance Rates Point To Need For More Equitable Heart Transplant Care
While access to donor hearts has increased, there are still gender- and race-based disparities in the acceptance rate of a donor heart offer by transplant teams, a new study finds.
STAT2 min read
STAT+: Pharmalittle: We’re Reading About Drug Shortages, Medicare Spending On An Alzheimer’s Drug, And More
Medicare estimates a new Alzheimer’s drug could cost the program billions of dollars by next year — well beyond what Wall Street or the drug’s manufacturer project.

Related Books & Audiobooks