A recent study has suggested that an artificial intelligence system is able to perform tasks far more accurately as opposed to a highly trained radiologist. The paper has been published in the Journal of the National Cancer Institute. It outlines how an artificial intelligence system can accurately detect cancer by evaluating digital mammography in breast cancer screenings.
Breast cancer screenings are an essential tool when it comes to early detection of breast cancer and the reduction of breast cancer-related deaths. Presently, screenings are quite labor intensive because of a large number of women requiring scans. In some parts of the globe, the US included, there exists an inadequacy of highly trained breast screening radiologists. This void has caused the development of artificial intelligence systems to help with tasks concerned with evaluations of mammograms.
Computer systems have been used for detecting and classifying breast lesions in mammograms autonomously since the 1980s. However, none of them have proven to be accurate enough to be put to clinical use. The recent study, however, shows that this paradigm is shifting. The study has compared the performance of a commercially available AI system against 101 radiologists that evaluated digital mammograms at a case level.
More than 28,000 interpretations were a part of the study. The artificial intelligence system was trained by making use of 9,000 mammograms that features cancer and another 180,000 mammograms that didn’t feature any abnormalities. The results showed that the AI system was statistically similar to the trained radiologists.
Loannis Sechopoulos, one of the paper’s authors, said, ‘Before we could decide what is the best way for AI systems to be introduced in the realm of breast cancer screening with mammography? We wanted to know how good can these systems really be. It was exciting to see that these systems have reached the level of matching the performance of not just radiologists, but of radiologists who spend at least a substantial portion of their time reading screening mammograms.’
Alejandro Rodriguez-Ruiz, MSc, Radboud University Medical Center in the Netherlands, writes, ‘Considering the increasing scarcity of radiologists in some countries, including breast screening radiologists, alternative strategies to allow continuation of current screening programs is required.’