Artificial intelligence (AI) can help radiologists distinguish false-positive mammograms from malignant and negative mammograms, according to new research published in Clinical Cancer Research.
“In order to catch breast cancer early and help reduce mortality, mammography is an important screening exam; however, it currently suffers from a high false recall rate,” co-author Shandong Wu, PhD, assistant professor of radiology, biomedical informatics, bioengineering, and clinical and translational science, and director of the Intelligent Computing for Clinical Imaging lab in the Department of Radiology at the University of Pittsburgh, Pennsylvania, said in a prepared statement from the American Association of Cancer Research. “These false recalls result in undue psychological stress for patients and a substantial increase in clinical workload and medical costs. Therefore, research on possible means to reduce false recalls in screening mammography is an important topic to investigate.”
Wu et al. constructed deep learning convolutional neural network (CNN) models to classify mammography images as malignant, negative or recalled-benign/false-positive. More than 14,000 images from more than 3,000 patients were used to train and test the CNN models.
Overall, using just images from a Full-Field Digital Mammography (FFDM) Dataset, the researchers obtained an AUC of 0.70-0.81. Using images from the FFDM Dataset and a Digital Dataset of Screening Mammography, the AUC was 0.77-0.96.
“We showed that there are imaging features unique to recalled-benign images that deep learning can identify and potentially help radiologists in making better decisions on whether a patient should be recalled or is more likely a false recall,” Wu said in the same prepared statement.
The algorithm’s ability to consistently categorize mammography data, Wu added, shows that AI “can augment radiologists in reading these images and ultimately benefit patients by helping reduce unnecessary recalls.”