Storage of magnetization as singlet order by optimal control designed pulses

Laustsen, Christoffer; Bowen, Sean; Vinding, Mads Sloth; Nielsen, Niels Chr.; Ardenkjær-Larsen, Jan Henrik

doi:10.1002/mrm.24733

Cited by 12 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1 We also propose to use DL for pulses designed with time-demanding or advanced techniques. This could be quantum mechanical approaches on multispin systems, 7,28,29 applications with very precise control updates, 30 or certain characteristics such as pulse envelope smoothness. 31 With the presented NNs, predicting a pulse takes on average about 7 ms on a laptop computer.…”

Section: Discussionmentioning

confidence: 99%

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

Vinding

Skyum

Sangill

et al. 2019

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

Purpose Some advanced RF pulses, like multidimensional RF pulses, are often long and require substantial computation time because of a number of constraints and requirements, sometimes hampering clinical use. However, the pulses offer opportunities of reduced‐FOV imaging, regional flip‐angle homogenization, and localized spectroscopy, e.g., of hyperpolarized metabolites. Proposed herein is a novel deep learning approach to ultrafast design of multidimensional RF pulses with intention of real‐time pulse updates. Methods The proposed neural network considers input maps of the desired excitation region of interest and outputs a single‐channel, multidimensional RF pulse. The training library is, e.g., retrieved from a large image database, and the target RF pulses trained upon are calculated with a method of choice. Results A relatively simple neural network is enough to produce reliable 2D spatial‐selective RF pulses of comparable performance to the teaching method. For binary regions of interest, the training library does not need to be vast; hence, reestablishment of the training library is not necessarily cumbersome. The predicted pulses were tested numerically and experimentally at 3 T. Conclusion Relatively effortless training of multidimensional RF pulses, based on non‐MRI‐related inputs, but working in an MRI setting still, has been demonstrated. The prediction time of a few milliseconds renders real‐time updates of advanced RF pulses possible.

show abstract

Section: Discussionmentioning

confidence: 99%

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

Vinding

Skyum

Sangill

et al. 2019

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

show abstract

“…Multidimensional spectral-/spatial-selective RF pulses have applications in inner-volume imaging, 1,2 navigation, 3,4 fMRI, 5,6 DWI, 7 DTI, 8 MRS, 9 carbon-13 dissolution dynamic nuclear polarization, 10,11 flow imaging, 12,13 and suppression. 14 RF pulse designers in general navigate between matters of pulse types; spatial (and/or spectral) target profiles; inhomogeneous B + 1 and B 0 fields; underlying gradient waveforms; system imperfections 15 ; constraints (eg, power, 16 specific absorption rate [SAR], 17 and states 18,19 ); and finally, which pulse design tool and physics model to adopt for the task…”

Section: Introductionmentioning

confidence: 99%

“…RF pulse designers in general navigate between matters of pulse types; spatial (and/or spectral) target profiles; inhomogeneous

B_{1}^{+}

and B 0 fields; underlying gradient waveforms; system imperfections 15 ; constraints (eg, power, 16 specific absorption rate [SAR], 17 and states 18,19 ); and finally, which pulse design tool and physics model to adopt for the task (eg, the Fourier‐based small tip angle regime, 20 a large tip angle optimal control type, 17,21,22 or a k‐space 23 or spatial domain algorithm) 24 …”

Section: Introductionmentioning

confidence: 99%

DeepControl: 2DRF pulses facilitating inhomogeneity and B₀ off‐resonance compensation in vivo at 7 T

Vinding

Aigner

Schmitter

et al. 2021

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

Funding information Kong Christian den Tiendes Fond; Harboefonden; Eva og Henry Fraenkels Mindefond; VILLUM FONDEN Purpose: Rapid 2DRF pulse design with subject-specific B + 1 inhomogeneity and B 0 off-resonance compensation at 7 T predicted from convolutional neural networks is presented. Methods: The convolution neural network was trained on half a million singlechannel transmit 2DRF pulses optimized with an optimal control method using artificial 2D targets, B + 1 and B 0 maps. Predicted pulses were tested in a phantom and in vivo at 7 T with measured B + 1 and B 0 maps from a high-resolution gradient echo sequence. Results: Pulse prediction by the trained convolutional neural network was done on the fly during the MR session in approximately 9 ms for multiple hand-drawn regions of interest and the measured B + 1 and B 0 maps. Compensation of B + 1 inhomogeneity and B 0 off-resonances has been confirmed in the phantom and in vivo experiments. The reconstructed image data agree well with the simulations using the acquired B + 1 and B 0 maps, and the 2DRF pulse predicted by the convolutional neural networks is as good as the conventional RF pulse obtained by optimal control. Conclusion: The proposed convolutional neural network-based 2DRF pulse design method predicts 2DRF pulses with an excellent excitation pattern and compensated B + 1 and B 0 variations at 7 T. The rapid 2DRF pulse prediction (9 ms) enables subjectspecific high-quality 2DRF pulses without the need to run lengthy optimizations. K E Y W O R D S 2DRF pulses, 7 T, artificial intelligence, deep learning, optimal control 1 | INTRODUCTION Multidimensional spectral-/spatial-selective RF pulses have applications in inner-volume imaging, 1,2 navigation, 3,4 fMRI, 5,6 DWI, 7 DTI, 8 MRS, 9 carbon-13 dissolution dynamic nuclear polarization, 10,11 flow imaging, 12,13 and suppression. 14 RF pulse designers in general navigate between matters of pulse types; spatial (and/or spectral) target profiles; inhomogeneous B + 1 and B 0 fields; underlying gradient waveforms; system imperfections 15 ; constraints (eg, power, 16 specific absorption rate [SAR], 17 and states 18,19); and finally, which pulse design tool and physics model to adopt for the task

show abstract

“…The decay time constants of such states are often many times the spinlattice relaxation time constant T 1 , with lifetimes exceeding 1 hour being observed in favourable cases 12,19 . Longlived states have been applied to the study of slow chemical and transport processes [20][21][22][23][24][25][26][27][28][29] , to the characterisation of biomolecular ligand binding [30][31][32][33][34] , and for the transport and storage of nuclear hyperpolarization 5,7,[35][36][37][38][39][40][41][42][43] . Longlived nuclear singlet order plays a central role in the generation of nuclear hyperpolarization from hydrogen gas enriched in the para spin isomer [44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59] .…”

Section: Introductionmentioning

confidence: 99%

Nuclear Singlet Relaxation by Chemical Exchange

Bengs

Dagys

Moustafa

et al. 2021

Preprint

View full text Add to dashboard Cite

The population imbalance between nuclear singlet states and triplet states of strongly coupled spin-1/2 pairs, also known as nuclear singlet order, is well protected against several common relaxation mechanisms. We study the nuclear singlet relaxation of 13C pairs in aqueous solutions of 1,2-13C2 squarate, over a range of pH values. The 13C singlet order is accessed by introducing 18O nuclei in order to break the chemical equivalence. The squarate dianion is in chemical equilibrium with hydrogen-squarate (SqH−) and squaric acid (SqH2) characterised by the dissociation constants pKa1 = 1:5 and pKa2 = 3:4. Surprisingly, we observe a striking increase in the singlet decay time constants TS when the pH of the solution exceeds ~ 10, which is far above the acid-base equilibrium points. We derive general rate expressions for chemical-exchange-induced nuclear singlet relaxation and provide a qualitative explanation of the TS behaviour of the squarate dianion. We identify a kinetic contribution to the singlet relaxation rate constant which depends explicitly on kinetic rate constants. Qualitative agreement is achieved between the theory and the experimental data. This study shows that infrequent chemical events may have a strong effect on the relaxation of nuclear singlet order.

show abstract

Storage of magnetization as singlet order by optimal control designed pulses

Abstract: It is demonstrated that significantly improved efficiency and robustness can be obtained within the limitations of typical MR scanner performance.

Cited by 12 publications

References 30 publications

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

DeepControl: 2DRF pulses facilitating inhomogeneity and B₀ off‐resonance compensation in vivo at 7 T

Nuclear Singlet Relaxation by Chemical Exchange

Contact Info

Product

Resources

About

Storage of magnetization as singlet order by optimal control designed pulses

Abstract: It is demonstrated that significantly improved efficiency and robustness can be obtained within the limitations of typical MR scanner performance.

Cited by 12 publications

References 30 publications

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

Ultrafast (milliseconds), multidimensional RF pulse design with deep learning

DeepControl: 2DRF pulses facilitating inhomogeneity and B0 off‐resonance compensation in vivo at 7 T

Nuclear Singlet Relaxation by Chemical Exchange

Contact Info

Product

Resources

About

DeepControl: 2DRF pulses facilitating inhomogeneity and B₀ off‐resonance compensation in vivo at 7 T