Oral presentation

Computer-aided detection (CAD) systems that attract the reader's attention to potentially suspicious features on a mammogram are now in clinical use in screening mammography in the USA and some European countries. However, recent publications have debated the benefit of CAD systems in screening mammography and have highlighted the need for robust evidence from prospective randomised trials. The evidence from these studies will be reviewed, including a recently published meta-analysis of double reading versus single reading with CAD.

The CADET II trial was a prospective multicentre randomised comparison of single reading with CAD and double reading in the UK National Health Service Breast Screening Programme. Over 30,000 women (age 50 to 70 years), attending routine mammo-graphy at three UK breast screening centres, were recruited into the trial. Film batches from screening sessions were randomly assigned in a ratio of 28:1:1 to one of three film-reading regimes: double reading and single reading with CAD, or double reading only, or single reading with CAD only. The primary outcome measures were matched comparisons of the cancer detection rates and the number of women recalled for assessment by the two reading regimes. The results of the study will be presented at the conference. The implications of this study will be discussed together with further work that needs to be undertaken.