Clipped DeepControl: Deep neural network two-dimensional pulse design with an amplitude constraint layer

TY - JOUR

T1 - Clipped DeepControl

T2 - Deep neural network two-dimensional pulse design with an amplitude constraint layer

AU - Vinding, Mads Sloth

AU - Lund, Torben Ellegaard

N1 - Copyright © 2022 The Author(s). Published by Elsevier B.V. All rights reserved.

PY - 2023/1

Y1 - 2023/1

N2 - Advanced radio-frequency pulse design used in magnetic resonance imaging has recently been demonstrated with deep learning of (convolutional) neural networks and reinforcement learning. For two-dimensionally selective radio-frequency pulses, the (convolutional) neural network pulse prediction time (a few milliseconds) was in comparison more than three orders of magnitude faster than the conventional optimal control computation. The network pulses were from the supervised training capable of compensating scan-subject dependent inhomogeneities of B 0 and B 1 + fields. Unfortunately, the network presented with a small percentage of pulse amplitude overshoots in the test subset, despite the optimal control pulses used in training were fully constrained. Here, we have extended the convolutional neural network with a custom-made clipping layer that completely eliminates the risk of pulse amplitude overshoots, while preserving the ability to compensate for the inhomogeneous field conditions.

AB - Advanced radio-frequency pulse design used in magnetic resonance imaging has recently been demonstrated with deep learning of (convolutional) neural networks and reinforcement learning. For two-dimensionally selective radio-frequency pulses, the (convolutional) neural network pulse prediction time (a few milliseconds) was in comparison more than three orders of magnitude faster than the conventional optimal control computation. The network pulses were from the supervised training capable of compensating scan-subject dependent inhomogeneities of B 0 and B 1 + fields. Unfortunately, the network presented with a small percentage of pulse amplitude overshoots in the test subset, despite the optimal control pulses used in training were fully constrained. Here, we have extended the convolutional neural network with a custom-made clipping layer that completely eliminates the risk of pulse amplitude overshoots, while preserving the ability to compensate for the inhomogeneous field conditions.

KW - Phantoms, Imaging

KW - Neural Networks, Computer

KW - Heart Rate

KW - Magnetic Resonance Imaging/methods

KW - Radio Waves

U2 - 10.1016/j.artmed.2022.102460

DO - 10.1016/j.artmed.2022.102460

M3 - Journal article

C2 - 36628795

SN - 0933-3657

VL - 135

JO - Artificial Intelligence in Medicine

JF - Artificial Intelligence in Medicine

M1 - 102460

ER -