Juhyung Park1, Dongmyung Shin1, Hyeong-Geol Shin1, Jiye Kim1, and Jongho Lee1
1Seoul National University, Seoul, Korea, Republic of
1Seoul National University, Seoul, Korea, Republic of
Slab-encoding RF pulses are designed
using deep reinforcement learning to optimize the condition number of the RF
encoding matrix. The newly designed RF pulses show improved robustness and
performance compared to conventional gSlider designs.
Figure 1. Overview of DeepSlider. Five
slab-encoding RF pulses are designed using deep reinforcement learning and
gradient descent. The signals
from slab-encoding RF pulses can be modeled as
a linear system of an encoding matrix
and sub-slice signals z. The design objective is
to minimize the condition number of the encoding matrix A, improving stability
against noise.
Figure 2. Details of generating
slab-encoding RF pulses. (a) In a single episode, DRL agent generates an RF
pulse as an action, and Bloch equation
environment returns a reward for the RF pulse. The reward is composed of slice
profile penalty and SAR regularization. The agent is updated using the reward
and generates an RF pulse recursively to maximize the reward. (b) DRL-generated
RF pulses are randomly grouped into five RF pulses. Then the five RF pulses are
optimized to have the lowest condition number in the encoding matrix
(and maximize DRL reward term) by gradient
descent.
