1Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea, Republic of, 2Department of Biomedical Engineering, Hankuk University of Foreign Studies, Gyeonggi-do, Korea, Republic of
In this study, we developed an approach of applying deep reinforcement learning for a RF pulse design in order to generate a machine-optimized RF pulse. Deep reinforcement learning was adopted to find the best root-flipping pattern for the minimum peak RF pulse in the SLR RF pulse design. When a multiband RF with high TBW/multiband factor was designed, the deep reinforcement learning showed much shorter duration than the modulated SLR RF pulse for the same RF peak constraint. Phantom and in-vivo scans were performed to demonstrate the feasibility of the newly designed RF pulse.
This abstract and the presentation materials are available to members only;
a login is required.
