Researchers in the United States and Britain have reported that a system of artificial intelligence in Google has proven similar accuracy to the accuracy of radiologists in the discovery of women who developed breast cancer based on mammograms and showed efficiency in reducing errors.
The study, published in the journal Nature, is the latest research showing that artificial intelligence has the potential to improve the accuracy of screening for breast cancer affecting a woman out of every eight women in the world.
According to the American Cancer Society, the percentage of error in radiologists in diagnosing breast cancer through X-ray imaging is 20 percent, and that half of the women who underwent examination over a period of ten years were mistakenly diagnosed with the disease.
One of the participants in preparing the study from Northwestern University School of Medicine, Muzyar Istadiadi, said that the results of the study, which was developed in cooperation with the (Deep Mind) unit for artificial intelligence in Google, which was integrated with Google Health in September, represented a significant progress in the possibility of early detection of breast cancer .
The team, which included researchers from the Imperial College London and the National Health Service in Britain, trained the system to identify breast cancer on tens of thousands of mammograms.
They then compared the system's expectations with the actual results for a group of 25,856 mammograms in Britain and 3,097 in the United States.
The study showed that the artificial intelligence system was able to determine the incidence of cancer to a degree similar to the accuracy of radiologists, while reducing the number of false positive results by 5.7 percent in the group in the United States and by 2.1 percent in those in Britain.
It also reduced false negative results, which tests were wrongly classified as normal, by 9.4 percent in the American group and 2.7 percent in the British group.
These differences reflect the ways to read the results of mammograms. In the United States, only one radiologist reads the results and takes tests every year or two.
In Britain, tests are taken every three years, and each expert is examined by radiologists. When they differ, a third expert is consulted.