A JOINT STUDY OF COMPUTER AIDED DETECTION (CAD) IN BREAST SCREENING
AT THE UNIVERSITY OF ABERDEEN, THE UNIVERSITY OF MANCHESTER &
QUEEN MARY COLLEGE, LONDON
Funded by CANCER RESEARCH UK and the NHS Breast Screening Programme
A study, which could potentially increase the detection rate of breast cancer by using the latest Computer Aided Detection system, has just been launched in the UK - the CADET Trial.
Teams of radiologists, scientists and radiographers from the University of Aberdeen, the University of Manchester and the South Manchester University Hospitals Trust will be using the R2 Technology ImageChecker to help detect potentially cancerous areas on mammograms.
The ImageChecker helps radiologists and radiographers in a similar manner to a PC spellchecker by automatically detecting and prompting suspicious areas on mammograms. It acts as a second pair of eyes and, therefore, could help to make mammogram checks more accurate.
It is hoped that the technology could reduce oversights and bring about earlier detection of cancerous lesions - both of which would reduce patient suffering and treatment costs.
The technology could also potentially alleviate the shortage of radiologists and radiographers which currently exists in the UK. At the moment, many mammograms are checked by two radiologists/radiographers, but the study will look at whether a single radiologist/radiographer prompted by the ImageChecker can produce the same or better results. If this is the case, savings would result.
The joint study is being led by Professor Fiona Gilbert, Head of Radiology at the University of Aberdeen, Dr Sue Astley, Senior Lecturer in Imaging Science at the University of Manchester, Professor Stephen Duffy, Professor of Cancer Screening, Queen Mary College, London and Dr Caroline Boggis, Consultant Radiologist at South Manchester University Hospitals Trust. The study is being funded by Cancer Research UK and the NHS Breast Screening Programme (NHSBSP).
Commenting on the study, Professor Gilbert, said: "Recent enhancements to the NHS Breast Screening Programme, including the requirement for additional mammogram views (Note 1) and an extended age range of women invited to screening (Note 2), coupled with a natural increase (Note 3) in the eligible screening population mean that screening volumes are set to increase dramatically.
"There is already a shortage of radiologists and radiographers and in the face of this burgeoning demand, we anticipate problems in delivering the service. This prompting technology needs to be examined in the NHSBSP setting, as it could potentially provide a solution to the shortage of film readers. The research team is confident of the importance and timeliness of this study."
Dr Astley from the University of Manchester added: "We will investigate whether individual radiologists, prompted by the ImageChecker, can match the level of detection currently considered to be the gold standard in the UK.
"Known as double reading with arbitration, this gold standard involves independent reading of each mammogram by two radiologists or radiographers with referral to a third radiologist if interpretation differs. We believe that the probability-coded prompt markers provided by the system could permit film readers to utilise the prompting information more effectively. We look forward to the results."
Dr Caroline Boggis from South Manchester University Hospitals Trust, commented: "This study will evaluate the contribution of R2/CAD to avoiding potential oversights. If effectiveness is established, the implication will be that the use of CAD will improve confidence in breast screening decisions."
The ImageChecker system digitises and analyses mammograms, displaying its findings on a screen adjacent to the mammogram viewer. Suspicious masses and micro calcifications are indicated by different symbols, sized according to the degree of concern.
The film reader first makes an unprompted assessment of the original mammogram, then consults the CAD prompt image and if appropriate, revises the assessment and decision to recall the patient or not, in light of the additional information.
For further information about the study and/or to arrange interviews with Professor Fiona Gilbert, Dr Sue Astley or Dr Caroline Boggis, contact:
Murray Fuller, Fuller:Anderson:Bellamy
Telephone: 020 7792 2302 or email: fab.partners@virgin.net
or
Angela Ferguson, Communications Team, Office of External Affairs, University of Aberdeen