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

arXiv: https://arxiv.org/abs/2201.08668 · HTML: https://mlsanity.com/paper/2201.08668/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 35 references that are in the map.
- First posted on arXiv 21 Jan 2022.

## Builds on (most cited first, 9 of 9)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: a review](https://mlsanity.com/paper/1712.03747/) (2017)
- [Deep neural networks are robust to weight binarization and other non-linear distortions](https://mlsanity.com/paper/1606.01981/) (2016)
- [Deep reinforcement learning-designed radiofrequency waveform in MRI](https://mlsanity.com/paper/2105.03061/) (2021)
- [Multi-structure bone segmentation in pediatric MR images with combined regularization from shape priors and adversarial network](https://mlsanity.com/paper/2009.07092/) (2020)
- [DeepControl: 2D RF pulses facilitating B1^+ inhomogeneity and B0 off-resonance compensation in vivo at 7T](https://mlsanity.com/paper/2009.12408/) (2020)
- [CapillaryNet: An Automated System to Quantify Skin Capillary Density and Red Blood Cell Velocity from Handheld Vital Microscopy](https://mlsanity.com/paper/2104.11574/) (2021)
- [Efficient high-resolution RF pulse design applied to simultaneous multi-slice excitation](https://mlsanity.com/paper/1702.07503/) (2017)
- [Optimal control gradient precision trade-offs: application to fast generation of DeepControl libraries for MRI](https://mlsanity.com/paper/2107.00933/) (2021)

## Abstract (arXiv)

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.