Jan 2, 2020

Google develops AI system that outperforms radiologists in detecting breast cancer

Jacob Knutson

Photo: Lyu Liang/VCG via Getty Images

Google Health developed an artificial intelligence system that can identify cases of breast cancer from mammograms more accurately than radiologists according to an international study, the Financial Times reports.

Why it matters: It's the latest example of how AI could improve early detection of diseases and reduce both false positives and false negatives diagnoses.

  • Dr. Dominic King, the UK lead for Google Health, said the results showed how AI could help screen harder-to-detect, early stage cancers.

By the numbers: Conventional mammogram screenings fail to detect about 1 in 5 breast cancer cases, according to the American Cancer Society.

  • More than half of all women are given a false positive diagnosis every 10 years, leading to unnecessary treatments that cost the U.S. more than $4 billion a year, according to FT.
  • The Google algorithm was trained and tested on de-identified images from almost 120,000 mammograms in the US and the UK.

Go deeper:

Go deeper

Caitlin Owens

AI's health care hype

Illustration: Sarah Grillo/Axios

Two new studies highlight artificial intelligence's potential to improve patient care, specifically by aiding or improving cancer detection.

Why it matters: AI could create enormous benefits for patients and the doctors who treat them, but some experts warn that the explosion of new health technology could put some patients in danger, as the L.A. Times and Kaiser Health News recently reported.

Go deeperArrowJan 7, 2020
Rashaan Ayesh

Cancer death rates drop by largest amount on record in U.S.

Photo: Harry Sieplinga/Getty Images

American Cancer Society researchers revealed in a new report published Wednesday that the U.S. cancer death rate dropped 2.2% between 2016 and 2017, the largest decline recorded in national cancer statistics dating back to 1930, AP reports.

The big picture via Axios' Bob Herman: Lung cancer drove most of the decline, as fewer people smoke cigarettes, and advanced lung cancer treatments become standard. Lung cancer accounts for nearly a quarter of all cancer deaths, according to the lead author of the report, Rebecca Siegel.

Go deeperArrowJan 8, 2020
Eileen Drage O'Reilly

Study suggests AI can help diagnose brain tumors

Three SRH images from biopsies: lymphoma, piilocytic astrocytomas brain tumor, and front lobal lesion (from l to r). Photo: Daniel Orringer/NYU Langone Health

In a clinical trial, a team of doctors found a combination of deep-learning algorithms and laser-imaging technology was as effective as pathologists — but faster — at diagnosing brain tumors in near real time, according to a study published Monday in Nature Medicine.

Why it matters: There is a shortage of pathologists and an increase in demand, with roughly 15.2 million people diagnosed with cancer each year globally, most of whom will have surgery. In the U.S. alone, there are roughly 1.1 million biopsies annually.

Go deeperArrowJan 7, 2020