Radiologists working in tandem with artificial intelligence more accurately detect breast cancer than either of them analyzing the images alone, a July study in The Lancet Digital Health found.
The study used a dataset of 1.2 million mammograms taken between 2007 and 2020 and was led by researchers from Vara, a German startup that offers the AI technology; Memorial Sloan Kettering Cancer Center in New York City; and the Medical University of Vienna in Austria.
When the AI system referred a suspicious mammogram to the radiologist, the findings were more precise than when either the technology or a human read them alone.
"This approach has the potential to improve the screening accuracy of radiologists, is adaptive to the requirements of screening, and could allow for the reduction of workload ahead of the consensus conference, without discarding the generalized knowledge of radiologists," the study's authors wrote.