Synthetic kidney ASL data with respiratory motion was generated using models from the XCAT phantom and matching recommendations for in-vivo acquisitions. Both pCASL and PASL datasets with 1 M0 and 25 control-label pairs were created and analysed using an in-house developed processing pipeline including registration, manual segmentation, calculation of mean perfusion-weighted image and perfusion map. The registration performed well on the synthetic data and the perfusion maps yielded good cortex/medulla contrast. The presented method allows a wide range of parameter choices for creating synthetic ASL datasets valuable for testing processing pipelines and comparing them across research and clinical imaging centres.
This abstract and the presentation materials are available to members only; a login is required.