Meeting Banner
Abstract #3658

Variation of Regions of Interest (ROIs) using different tools for automatic ROI generation: The impact on reported magnetic susceptibility values in QSM

Emma Dixon1, David L Thomas2, Anna Barnes3, and Karin Shmueli1

1Medical Physics and Biomedical Engineering, University College London, London, United Kingdom, 2Leonard Wolfson Experimental Neurology Centre, University College London, Institute of Neurology, United Kingdom, 3Institute of Nuclear Medicine, UCLH-NHS Foundation Trust, United Kingdom

Automatic Region of Interest (ROI) generation is useful for large clinical studies using Quantitative Susceptibility Mapping (QSM) as this allows mean susceptibility values for anatomical regions to be reported without manual intervention.

Several methods to generate ROIs are compared in this work. We found that methods based on T1 contrast showed little variation whether a typical magnitude image from a susceptibility protocol or a structural MPRAGE image was used. An additional method based on a QSM atlas showed greater variability in ROIs.

Mean susceptibility values for ROIs were shown to vary greatly depending on the method used to generate ROIs.

This abstract and the presentation materials are available to members only; a login is required.

Join Here