It’s a well-established fact that mammography reduces breast cancer mortality. The high rate of false-positive recalls associated with alternative screenings has accelerated the development of AI-driven systems from , , and elsewhere. But they aren’t perfect, because most models operate on a single screening exam as opposed to more recent exams.
This shortcoming motivated a team of researchers at New York University’s Center for Data Science and Department of Radiology to propose (“Screening Mammogram Classification with Prior Exams “) a machine learning framework that takes advantage of prior exams in making a diagnosis. They say that in preliminary tests, it reduced the error rate of the baseline and achieved an area under the curve (a metric indicating performance at all classification thresholds) of 0.8664 for