Meeting Banner
Abstract #0345

Deep Learning Predicts Synthetic T1rho Maps from T2 Maps for Knee MRI

Michelle W. Tong1,2, Aniket A. Tolpadi1,2, Alex Beltran1,2, Sharmila Majumdar2, and Valentina Pedoia 2
1Department of Bioengineering, University of California Berkeley, Berkeley, CA, United States, 2Department of Radiology and Biomedical Imaging, University of California San Francsico, San Francisco, CA, United States


This study explores the use a deep learning network to generate T1rho maps from T2 maps for knee MRI. T1rho maps have clinical value in the diagnosis of osteoarthritis in addition to T2 maps while T2 maps are more widely adopted and available in clinical datasets. This study found synthetic T1rho maps images maintain excellent fidelity for data collected in a research setting while performance is reduced for data collected in a clinical setting. This work elucidates the promise of deep learning in accelerating imaging protocols through domain adaptation as opposed to more common reconstruction approaches.

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

Join Here