A study, which was published in Nature, explored how the AI diagnoses compared to physician diagnoses. In the study, Google “trained” a computer to find abnormalities on X-ray images of approximately 29,000 women. For the study, six U.S. radiolgists were given 500 mammograms to interpret. The radiologists’ findings were compared to the findings of the AI. The physicians were also given information on the patients, including age, breast cancer history, and previous screening mammograms.
Among U.S. patients, the AI reduced the number of false positives, where a mammogram incorrectly diagnoses an X-ray as abnormal, by 5.7 percent. Additionally, there was a 9.4 percent reduction in false negatives, which is when the cancer goes undetected.
“In an independent study of six radiologists, the AI system outperformed all of the human readers,” the study claimed, according to CNBC.
To access the complete study, click here.
More articles on AI:
Rushing unregulated AI into healthcare raises patient risks: ‘I don’t think we want to fail fast’
8 healthcare companies among the top 25 highest-funded AI startups of 2019
NAM report: The success of healthcare AI depends on building consensus
