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

TL;DR

This paper introduces Clipped DeepControl, a neural network-based method for 2D pulse design in MRI that ensures amplitude constraints, significantly improving safety and reliability over previous models.

Contribution

It develops a novel clipping layer for neural networks that prevents pulse amplitude overshoots while maintaining compensation for field inhomogeneities.

Findings

The clipping layer effectively eliminates amplitude overshoots.

The method maintains fast prediction times comparable to previous neural networks.

It improves safety and reliability in MRI pulse design.

Abstract

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 (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 B0 and B+1 fields. Unfortunately, the network presented with a non-negligible 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 the inhomogeneous field conditions.