1
|
Kilic T, Liebig P, Demirel OB, Herrler J, Nagel AM, Ugurbil K, Akçakaya M. Unsupervised deep learning with convolutional neural networks for static parallel transmit design: A retrospective study. Magn Reson Med 2024; 91:2498-2507. [PMID: 38247050 PMCID: PMC10997461 DOI: 10.1002/mrm.30014] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/07/2023] [Revised: 12/28/2023] [Accepted: 12/29/2023] [Indexed: 01/23/2024]
Abstract
PURPOSE To mitigateB 1 + $$ {B}_1^{+} $$ inhomogeneity at 7T for multi-channel transmit arrays using unsupervised deep learning with convolutional neural networks (CNNs). METHODS Deep learning parallel transmit (pTx) pulse design has received attention, but such methods have relied on supervised training and did not use CNNs for multi-channelB 1 + $$ {B}_1^{+} $$ maps. In this work, we introduce an alternative approach that facilitates the use of CNNs with multi-channelB 1 + $$ {B}_1^{+} $$ maps while performing unsupervised training. The multi-channelB 1 + $$ {B}_1^{+} $$ maps are concatenated along the spatial dimension to enable shift-equivariant processing amenable to CNNs. Training is performed in an unsupervised manner using a physics-driven loss function that minimizes the discrepancy of the Bloch simulation with the target magnetization, which eliminates the calculation of reference transmit RF weights. The training database comprises 3824 2D sagittal, multi-channelB 1 + $$ {B}_1^{+} $$ maps of the healthy human brain from 143 subjects.B 1 + $$ {B}_1^{+} $$ data were acquired at 7T using an 8Tx/32Rx head coil. The proposed method is compared to the unregularized magnitude least-squares (MLS) solution for the target magnetization in static pTx design. RESULTS The proposed method outperformed the unregularized MLS solution for RMS error and coefficient-of-variation and had comparable energy consumption. Additionally, the proposed method did not show local phase singularities leading to distinct holes in the resulting magnetization unlike the unregularized MLS solution. CONCLUSION Proposed unsupervised deep learning with CNNs performs better than unregularized MLS in static pTx for speed and robustness.
Collapse
Affiliation(s)
- Toygan Kilic
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, Minnesota, USA
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA
| | | | - Omer Burak Demirel
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, Minnesota, USA
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA
| | | | - Armin M Nagel
- Institute of Radiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany
- Division of Medical Physics in Radiology, German Cancer Research Centre (DKFZ), Heidelberg, Germany
| | - Kamil Ugurbil
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA
| | - Mehmet Akçakaya
- Electrical and Computer Engineering, University of Minnesota, Minneapolis, Minnesota, USA
- Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA
| |
Collapse
|
2
|
Kopanoglu E. Actual patient position versus safety models: Specific Absorption Rate implications of initial head position for Ultrahigh Field Magnetic Resonance Imaging. NMR Biomed 2023; 36:e4876. [PMID: 36385447 PMCID: PMC10802886 DOI: 10.1002/nbm.4876] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/08/2022] [Revised: 10/20/2022] [Accepted: 11/13/2022] [Indexed: 06/16/2023]
Abstract
Specific absorption rate (SAR) relates power absorption to tissue heating, and therefore is used as a safety constraint in magnetic resonance imaging (MRI). This study investigates the implications of initial head positioning on local and whole-head SAR. A virtual body model was simulated at 161 positions inside an eight-channel parallel-transmit (pTx) array. On-axis displacements and rotations of up to 20 mm/degrees and off-axis axial/coronal translations were investigated. Single-channel, radiofrequency (RF) shimming (i.e., single-spoke pTx) and multispoke pTx pulses were designed for seven axial, five coronal and five sagittal slices at each position (the slices were consistent across all positions). Whole-head and local SAR were calculated using safety models consisting of a single (centred) body position, multiple representative positions and all simulated body positions. Positional mismatches between safety models and actual positions cause SAR underestimation. For axial imaging, the actual peak local SAR was up to 4.2-fold higher for both single-channel and 5-spoke pTx, 3.5-fold higher for 3-/4-spoke pTx, and 2-fold higher for RF shimming and 2-spoke pTx, compared with that calculated using the centred body position. For sagittal and coronal imaging, the underestimation of peak local SAR was up to 5.2-fold and 3.8-fold, respectively. Using all body positions to estimate SAR prevented SAR underestimation but yielded up to 11-fold SAR overestimation for RF shimming. Local SAR of single-channel and pTx multispoke pulses showed considerable dependence on the initial patient position. RF shimming yielded much lower sensitivity to positional mismatches for axial imaging but not for sagittal and coronal imaging. This was deemed attributable to the higher degrees-of-freedom of control offered by the investigated coil array for axial imaging. Whole-head SAR is less sensitive to positional mismatches compared with local SAR. Nevertheless, whole-head SAR increased by up to 80% for sagittal imaging. Local and whole-head SAR were observed to be more sensitive to positional mismatches in the axial plane, because of larger variations in coil-tissue proximity. Using all possible body positions in the safety model may become substantially over-conservative and limit imaging performance, especially for the RF shimming mode for axial imaging.
Collapse
Affiliation(s)
- Emre Kopanoglu
- Cardiff University Brain Research Imaging Centre (CUBRIC), School of PsychologyCardiff UniversityCardiffUK
| |
Collapse
|
3
|
Herrler J, Williams SN, Liebig P, Ding B, McElhinney P, Allwood-Spiers S, Meixner CR, Gunamony S, Maier A, Dörfler A, Gumbrecht R, Porter DA, Nagel AM. The effects of RF coils and SAR supervision strategies for clinically applicable nonselective parallel-transmit pulses at 7 T. Magn Reson Med 2023; 89:1888-1900. [PMID: 36622945 DOI: 10.1002/mrm.29569] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/02/2022] [Revised: 12/13/2022] [Accepted: 12/14/2022] [Indexed: 01/10/2023]
Abstract
PURPOSE To investigate the effects of using different parallel-transmit (pTx) head coils and specific absorption rate (SAR) supervision strategies on pTx pulse design for ultrahigh-field MRI using a 3D-MPRAGE sequence. METHODS The PTx universal pulses (UPs) and fast online-customized (FOCUS) pulses were designed with pre-acquired data sets (B0 , B1 + maps, specific absorption rate [SAR] supervision data) from two different 8 transmit/32 receive head coils on two 7T whole-body MR systems. For one coil, the SAR supervision model consisted of per-channel RF power limits. In the other coil, SAR estimations were done with both per-channel RF power limits as well as virtual observation points (VOPs) derived from electromagnetic field (EMF) simulations using three virtual human body models at three different positions. All pulses were made for nonselective excitation and inversion and evaluated on 132 B0 , B1 + , and SAR supervision datasets obtained with one coil and 12 from the other. At both sites, 3 subjects were examined using MPRAGE sequences that used UP/FOCUS pulses generated for both coils. RESULTS For some subjects, the UPs underperformed when simulated on a different coil from which they were derived, whereas FOCUS pulses still showed acceptable performance in that case. FOCUS inversion pulses outperformed adiabatic pulses when scaled to the same local SAR level. For the self-built coil, the use of VOPs showed reliable overestimation compared with the ground-truth EMF simulations, predicting about 52% lower local SAR for inversion pulses compared with per-channel power limits. CONCLUSION FOCUS inversion pulses offer a low-SAR alternative to adiabatic pulses and benefit from using EMF-based VOPs for SAR estimation.
Collapse
Affiliation(s)
- Jürgen Herrler
- Department of Neuroradiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.,Siemens Healthcare, Erlangen, Germany
| | | | | | | | - Paul McElhinney
- Imaging Center of Excellence, University of Glasgow, Glasgow, UK
| | | | - Christian R Meixner
- Siemens Healthcare, Erlangen, Germany.,Institute of Radiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
| | - Shajan Gunamony
- Imaging Center of Excellence, University of Glasgow, Glasgow, UK.,MR CoilTech, Glasgow, UK
| | - Andreas Maier
- Department of Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
| | - Arnd Dörfler
- Department of Neuroradiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
| | | | - David A Porter
- Imaging Center of Excellence, University of Glasgow, Glasgow, UK
| | - Armin M Nagel
- Institute of Radiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.,Division of Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany
| |
Collapse
|
4
|
Krueger F, Aigner CS, Hammernik K, Dietrich S, Lutz M, Schulz-Menger J, Schaeffter T, Schmitter S. Rapid estimation of 2D relative B 1 + -maps from localizers in the human heart at 7T using deep learning. Magn Reson Med 2023; 89:1002-1015. [PMID: 36336877 DOI: 10.1002/mrm.29510] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/15/2022] [Revised: 10/11/2022] [Accepted: 10/11/2022] [Indexed: 11/09/2022]
Abstract
PURPOSE Subject-tailored parallel transmission pulses for ultra-high fields body applications are typically calculated based on subject-specific B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps of all transmit channels, which require lengthy adjustment times. This study investigates the feasibility of using deep learning to estimate complex, channel-wise, relative 2D B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps from a single gradient echo localizer to overcome long calibration times. METHODS 126 channel-wise, complex, relative 2D B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps of the human heart from 44 subjects were acquired at 7T using a Cartesian, cardiac gradient-echo sequence obtained under breath-hold to create a library for network training and cross-validation. The deep learning predicted maps were qualitatively compared to the ground truth. Phase-only B 1 + $$ {\mathrm{B}}_1^{+} $$ -shimming was subsequently performed on the estimated B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps for a region of interest covering the heart. The proposed network was applied at 7T to 3 unseen test subjects. RESULTS The deep learning-based B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps, derived in approximately 0.2 seconds, match the ground truth for the magnitude and phase. The static, phase-only pulse design performs best when maximizing the mean transmission efficiency. In-vivo application of the proposed network to unseen subjects demonstrates the feasibility of this approach: the network yields predicted B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps comparable to the acquired ground truth and anatomical scans reflect the resulting B 1 + $$ {\mathrm{B}}_1^{+} $$ -pattern using the deep learning-based maps. CONCLUSION The feasibility of estimating 2D relative B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps from initial localizer scans of the human heart at 7T using deep learning is successfully demonstrated. Because the technique requires only sub-seconds to derive channel-wise B 1 + $$ {\mathrm{B}}_1^{+} $$ -maps, it offers high potential for advancing clinical body imaging at ultra-high fields.
Collapse
Affiliation(s)
- Felix Krueger
- Physikalisch-Technische Bundesanstalt, Braunschweig and Berlin, Germany.,Technische Universität Berlin, Biomedical Engineering, Berlin, Germany
| | | | - Kerstin Hammernik
- Technical University of Munich, Munich, Germany.,Imperial College London, London, United Kingdom
| | | | - Max Lutz
- Physikalisch-Technische Bundesanstalt, Braunschweig and Berlin, Germany
| | - Jeanette Schulz-Menger
- Charité-Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität Zu Berlin, Experimental Clinical Research Center, Berlin, Germany.,DZHK (German Centre for Cardiovascular Research), Partner Site Berlin, Berlin, Germany.,Department of Cardiology and Nephrology, HELIOS Hospital Berlin-Buch, Berlin, Germany
| | - Tobias Schaeffter
- Physikalisch-Technische Bundesanstalt, Braunschweig and Berlin, Germany.,Technische Universität Berlin, Biomedical Engineering, Berlin, Germany.,Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom
| | - Sebastian Schmitter
- Physikalisch-Technische Bundesanstalt, Braunschweig and Berlin, Germany.,Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA.,Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany
| |
Collapse
|
5
|
Williams SN, McElhinney P, Gunamony S. Ultra-high field MRI: parallel-transmit arrays and RF pulse design. Phys Med Biol 2023; 68. [PMID: 36410046 DOI: 10.1088/1361-6560/aca4b7] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/28/2022] [Accepted: 11/21/2022] [Indexed: 11/22/2022]
Abstract
This paper reviews the field of multiple or parallel radiofrequency (RF) transmission for magnetic resonance imaging (MRI). Currently the use of ultra-high field (UHF) MRI at 7 tesla and above is gaining popularity, yet faces challenges with non-uniformity of the RF field and higher RF power deposition. Since its introduction in the early 2000s, parallel transmission (pTx) has been recognized as a powerful tool for accelerating spatially selective RF pulses and combating the challenges associated with RF inhomogeneity at UHF. We provide a survey of the types of dedicated RF coils used commonly for pTx and the important modeling of the coil behavior by electromagnetic (EM) field simulations. We also discuss the additional safety considerations involved with pTx such as the specific absorption rate (SAR) and how to manage them. We then describe the application of pTx with RF pulse design, including a practical guide to popular methods. Finally, we conclude with a description of the current and future prospects for pTx, particularly its potential for routine clinical use.
Collapse
Affiliation(s)
- Sydney N Williams
- Imaging Centre of Excellence, University of Glasgow, Glasgow, United Kingdom
| | - Paul McElhinney
- Imaging Centre of Excellence, University of Glasgow, Glasgow, United Kingdom
| | - Shajan Gunamony
- Imaging Centre of Excellence, University of Glasgow, Glasgow, United Kingdom.,MR CoilTech Limited, Glasgow, United Kingdom
| |
Collapse
|