Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

2022 ◽  
Author(s):  
Sumit Tewari ◽  
Sahar Yousefi ◽  
Andrew G Webb
Keyword(s):  
Neural Network ◽  
Magnetic Resonance Imaging ◽  
Genetic Algorithm ◽  
Deep Neural Network ◽  
Training Dataset ◽  
Resonance Imaging ◽  
Magnet Design ◽  
Low Field ◽  
Magnetic Resonance Imaging Mri ◽  
Genetic Algorithm Model

Abstract We present a combination of a CNN-based encoder with an analytical forward map for solving inverse problems. We call it an encoder-analytic (EA) hybrid model. It does not require a dedicated training dataset and can train itself from the connected forward map in a direct learning fashion. A separate regularization term is not required either, since the forward map also acts as a regularizer. As it is not a generalization model it does not suffer from overfitting. We further show that the model can be customized to either finding a specific target solution or one that follows a given heuristic. As an example, we apply this approach to the design of a multi-element surface magnet for low-field magnetic resonance imaging (MRI). We further show that the EA model can outperform the benchmark genetic algorithm model currently used for magnet design in MRI, obtaining almost 10 times better results.

scholarly journals Effectiveness of low-field magnetic resonance imaging in diagnosing brachial plexus tumours in dogs – short communication

2015 ◽  
Vol 59 (2) ◽  
pp. 317-319
Author(s):  
Zbigniew Adamiak ◽  
Yauheni Zhalniarovich ◽  
Paulina Przyborowska ◽  
Joanna Głodek ◽  
Adam Przeworski
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Brachial Plexus ◽  
Resonance Imaging ◽  
Gadolinium Contrast ◽  
Gadolinium Contrast Agent ◽  
Low Field ◽  
Mri Sequences ◽  
Magnetic Resonance Imaging Mri ◽  
Brachial Plexus Tumours

AbstractThe aim of the study was to identify magnetic resonance imaging (MRI) sequences that contribute to a quick and reliable diagnosis of brachial plexus tumours in dogs. The tumours were successfully diagnosed in 6 dogs by the MRI with the use of SE, FSE, STIR, Turbo 3 D, 3D HYCE, and GE sequences and the gadolinium contrast agent

scholarly journals Diagnosa Tumor Otak Berdasarkan Citra MRI (Magnetic Resonance Imaging)

2019 ◽  
Vol 18 (2) ◽  
Author(s):  
Ida Bagus Leo Mahadya Suta ◽  
Rukmi Sari Hartati ◽  
Yoga Divayana
Keyword(s):  
Neural Network ◽  
Magnetic Resonance Imaging ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Convolutional Neural Network ◽  
Resonance Imaging ◽  
Magnetic Resonance Imaging Mri

Tumor otak menjadi salah satu penyakit yang paling mematikan, salah satu jenis yang paling banyak ditemukan adalah glioma sekitar 6 dari 100.000 pasien adalah penderita glioma. Citra digital melalui Magnetic Resonance Imaging (MRI) merupakan salah satu metode untuk membantu dokter dalam menganalisa dan mengklasifikasikan jenis tumor otak. Namun, klasifikasi secara manual membutuhkan waktu yang lama dan memiliki resiko kesalahan yang tinggi, untuk itu dibutuhkan suatu cara otomatis dan akurat dalam melakukan klasifikasi citra MRI. Convolutional Neural Network (CNN) menjadi salah satu solusi dalam melakukan klasifikasi otomatis dalam citra MRI. CNN merupakan algoritma deep learning yang memiliki kemampuan untuk belajar sendiri dari kasus kasus sebelumnya. Dan dari penelitian yang telah dilakukan, diperoleh hasil bahwa CNN mampu dalam menyelesaikan klasifikasi tumor otak dengan akurasi yang tinggi. Peningkatan akurasi diperoleh dengan mengembangkan algoritma CNN baik melalui menentukan nilai kernel dan/atau fungsi aktivasi.

Monitoring a typical fermentation process of natto by low-field nuclear magnetic resonance (LF-NMR) and magnetic resonance imaging (MRI) techniques

2016 ◽  
Vol 8 (39) ◽  
pp. 7135-7140 ◽  
Author(s):  
Jing Wu ◽  
Yanru Li ◽  
Xingsheng Gao
Keyword(s):  
Magnetic Resonance Imaging ◽  
Nuclear Magnetic Resonance ◽  
Magnetic Resonance ◽  
Fermentation Process ◽  
Resonance Imaging ◽  
Low Field ◽  
Magnetic Resonance Imaging Mri ◽  
Lf Nmr ◽  
Nuclear Magnetic

Unique insights into the monitoring of a typical fermentation process of natto based on low field nuclear magnetic resonance (LF-NMR) associated with magnetic resonance imaging (MRI).

Evaluation of Magnetic Resonance Imaging Responsiveness in Active Psoriatic Arthritis at Multiple Timepoints during the First 12 Weeks of Antitumor Necrosis Factor Therapy

2015 ◽  
Vol 43 (1) ◽  
pp. 75-80 ◽  
Author(s):  
Marie Feletar ◽  
Stephen Hall ◽  
Paul Bird
Keyword(s):  
Magnetic Resonance Imaging ◽  
Psoriatic Arthritis ◽  
Magnetic Resonance ◽  
High Field ◽  
Early Response ◽  
Resonance Imaging ◽  
Low Field ◽  
Low Field Mri ◽  
Magnetic Resonance Imaging Mri ◽  
Necrosis Factor

Objective.To assess the responsiveness of high- and low-field extremity magnetic resonance imaging (MRI) variables at multiple timepoints in the first 12 weeks post-antitumor necrosis factor (anti-TNF) therapy initiation in patients with psoriatic arthritis (PsA) and active dactylitis.Methods.Twelve patients with active PsA and clinical evidence of dactylitis involving at least 1 digit were recruited. Patients underwent sequential high-field conventional (1.5 Tesla) and extremity low-field MRI (0.2 Tesla) of the affected hand or foot, pre- and postgadolinium at baseline (pre-TNF), 2 weeks (post-TNF), 6 weeks, and 12 weeks. A blinded observer scored all images on 2 occasions using the PsA MRI scoring system.Results.Eleven patients completed the study, but only 6 patients completed all high-field and low-field MRI assessments. MRI scores demonstrated rapid response to TNF inhibition with score reduction in tenosynovitis, synovitis, and osteitis at 2 weeks. Intraobserver reliability was good to excellent for all variables. High-field MRI demonstrated greater sensitivity to tenosynovitis, synovitis, and osteitis and greater responsiveness to change posttreatment. Treatment responses were maintained to 12 weeks.Conclusion.This study demonstrates the use of MRI in detecting early response to biologic therapy. MRI variables of tenosynovitis, synovitis, and osteitis demonstrated responsiveness posttherapy with high-field scores more responsive to change than low-field scores.

scholarly journals B0-Shimming Methodology for Affordable and Compact Low-Field Magnetic Resonance Imaging Magnets

2021 ◽  
Vol 9 ◽  
Author(s):  
Konstantin Wenzel ◽  
Hazem Alhamwey ◽  
Tom O’Reilly ◽  
Layla Tabea Riemann ◽  
Berk Silemek ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Field ◽  
Magnetic Resonance ◽  
Point Of Care ◽  
Low Cost ◽  
Field Measurements ◽  
Target Volume ◽  
Resonance Imaging ◽  
Low Field ◽  
Magnetic Resonance Imaging Mri

Low-field (B0 < 0.2 T) magnetic resonance imaging (MRI) is emerging as a low cost, point-of-care alternative to provide access to diagnostic imaging technology even in resource scarce environments. MRI magnets can be constructed based on permanent neodymium-iron-boron (NdFeB) magnets in discretized arrangements, leading to substantially lower mass and costs. A challenge with these designs is, however, a good B0 field homogeneity, which is needed to produce high quality images free of distortions. In this work, we describe an iterative approach to build a low-field MR magnet based on a B0-shimming methodology using genetic algorithms. The methodology is tested by constructing a small bore (inner bore diameter = 130 mm) desktop MR magnet (<15 kg) at a field strength of B0 = 0.1 T and a target volume of 4 cm in diameter. The configuration consists of a base magnet and shim inserts, which can be placed iteratively without modifying the base magnet assembly and without changing the inner dimensions of the bore or the outer dimensions of the MR magnet. Applying the shims, B0 field inhomogeneity could be reduced by a factor 8 from 5,448 to 682 ppm in the target central slice of the magnet. Further improvements of these results can be achieved in a second or third iteration, using more sensitive magnetic field probes (e.g., nuclear magnetic resonance based magnetic field measurements). The presented methodology is scalable to bigger magnet designs. The MR magnet can be reproduced with off-the-shelf components and a 3D printer and no special tools are needed for construction. All design files and code to reproduce the results will be made available as open source hardware.

Sign in / Sign up

Export Citation Format

Share Document