Imaging studies

Research involving imaging modalities presents particular methodological challenges. These include decisions such as:

• What aspect of the image to measure
• What summary statistic to use
• How to deal with large amounts of multiple testing
• How to address possible diagnostic error

The main aim of this research section is to provide a networking group for statistical researchers involved in imaging studies in healthcare research. We meet twice-yearly and advertise our events on the main NIHR Statistics Group email list and allstat. We also write publication articles describing statistical issues that arise in imaging studies.

A major recent initiative has been the PROVIDENT considerations for clinical trials and studies using imaging. This brought together 30 professionals with experience of imaging studies to highlight the unique logistical challenges posed by imaging in research. A series of workshops developed the outline of the considerations, including input from:

• Academic and commercial researchers
• Radiologists, radiographers & medical physicists
• Clinicians & research nurses
• Patient advocates
• Trial managers, health economists, methodologists, statisticians, and
• People with central and individual site perspectives

We are now moving onto the next phase of this project, which will focus on the methodological challenges.

Our working group committee coordinates meetings and other activities. We are open to suggestions for topics and presentations for future meetings. Meetings to date have been organised to include 25 to 30 statisticians and researchers, enabling good networking and small group discussions.

If you would like to join a working group committee or wish to find out more, please email the NIHR Statistics Group.