A Google artificial intelligence system bested 6 radiologists in identifying whether a patient had lung cancer according to a new Nature Medicine study reported by Stat News. When looking at a particular CT scan, the algorithm detected 5% more cancers than the humans and reduced false positives by 11%. When previous images of patients were included in the evaluation, the algorithm scored the same as the radiologists.
Why this matters: Conducting screening on high-risk patients for lung cancer ultimately reduces the risk of death, however false positives can lead to a host of other issues including harmful and unnecessary surgeries.
Notable quotes: “These people have a technology that will improve the precision of screening tremendously.” Dr. Otis Brawley, a professor of oncology and epidemiology at Johns Hopkins University and formerly executive vice president of the American Cancer Society
“We were able to train the [AI models] over the whole scan at a very high resolution. Although radiologists can look at the slices, there are clear advantages to the model here.” -Shravya Shetty, Google engineer and technical lead on the study