Meeting Banner
Abstract #0971

Accelerating clinical diffusion-weighted MRI using deep learning: Potential utility in metastatic prostate cancer and malignant mesothelioma

Konstantinos Zormpas-Petridis1, Nina Tunariu1, Andra Curcean1, Christina Messiou1, David Collins1, Yann Jamin1, Dow-Mu Koh1, and Matthew D. Blackledge1
1Radiotherapy and Imaging, Institute of Cancer Research, London, Sutton, United Kingdom

Diffusion-weighted MR-imaging (DWI) is an attractive non-invasive tool for staging and response evaluation of myeloma and metastatic bone disease. However, scans can last up to 30 minutes in whole body studies, which can hinder the adoption of DWI in clinical practice, especially in patients who are unwell. Here, we use a deep learning approach to establish that sub-sampled, but rapidly acquired images, could be used to reconstruct ‘clinical-grade’ DWI images, potentially reducing acquisition times (from ~30 to ~5 minutes). Such time savings could reduce scanning costs and spare patient time/discomfort.

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

Join Here