Harnessing the benefits of AI in the breast imaging pathway

Author(s): Dr Richard Sidebottom, Dr Kevin Dunbar, Dr Louise Wilkinson

Hospital: Royal Marsden NHS Foundation Trust; NHS England and Improvement; Oxford University Hospitals NHS Foundation Trust

Reference: RAD Magazine, 48, 560, 19-20


The nature of modern radiology has resulted in extremely large volumes of images stored in local archives. This, combined with advances in computing power and improvements in machine learning techniques, means that AI systems are being developed that are powerful enough to be clinically valuable.

In the UK breast screening programme, image data can be matched with outcome data at an individual level, providing an ideal environment for training AI systems for image classification in the breast screening workflow. This is hailed by some as the answer to the shortage of breast radiologists and provides hope of ensuring greater consistency in the detection of significant abnormalities on mammograms and individual risk stratification. The high priority for policy makers and researchers is reflected in the allocation of two high value NHSX AI awards to companies that have validated algorithms for mammography interpretation.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read more