10 Clues Suggest Size, Shape of AI’s Future in Mammography

Computer-aided detection, supported by AI, has often proven superior to traditional CAD over the past decade, but the ‘new way’ has slowly gained wide acceptance.

The slowdown is largely the result of high entry costs coupled with difficult ethical and legal concerns. The sooner breast radiology works out the principles and specifics of these issues, the better breast imaging researchers will be at optimizing patient care.

The promise remains worth pursuing as AI-based CAD tools have been shown to reduce recalls and interval cancers, while streamlining workflows by automatically prioritizing negative mammograms.

So researchers from Columbia University who reviewed the literature on AI in mammography and had their report conclude[1] published on May 15 in Clinical Imaging

Richard Ha, MD, and Meghan Jairam, MD present illustrative findings from more than 30 published studies by summarizing recent advances in AI-assisted mammography and considering where momentum points to the future of AI in the field.

Among the gems they bring out are 10 key findings in five categories:

Leave a Comment