Pith. sign in

REVIEW 4 major objections 6 minor 123 references

Hardware Acceleration in Portable MRIs: State of the Art and Future Prospects

T0 review · 4 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This review argues that hardware acceleration is essential for portable MRI viability and calls for shared low-field datasets and a three-stage evidence ladder to get AI-based reconstruction from prototype to regulatory and clinical maturit

desk verdict Useful survey of an under-reviewed niche, but the conclusion overstates the evidence: acceleration is promising, not demonstrated essential, for pMRI. read the letter →

arxiv 2509.06365 v1 pith:VV67NKHW submitted 2025-09-08 cs.AR

Omar Al Habsi , Safa Mohammed Sali , Anis Meribout , Mahmoud Meribout , Saif Almazrouei , Mohamed Seghier This is my paper
classification cs.AR
keywords portableMRIlow-fieldhardwareaccelerationFPGAGPUASICdeeplearningreconstructionpoint-of-careimaging
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This review argues that hardware acceleration is not an optional extra but a necessary component of viable portable MRI: without edge GPUs, FPGAs, or ASICs, the reconstruction and AI workloads that compensate for low-field noise and undersampling cannot run fast enough or efficiently enough for point-of-care use. The authors surveyed current pMRI prototypes and accelerator results, concluding that FPGAs currently offer the best balance of latency, power, and reconfigurability for portable systems, while GPUs dominate research prototyping and ASICs promise the highest efficiency at high volume. They also propose a Low-Field MRI Consortium and a three-step evidence ladder—phantom validation, retrospective multi-center testing, and prospective reader and non-inferiority trials—to supply the public datasets and regulator-ready benchmarks that AI-based low-field reconstruction currently lacks. If the claim holds, future pMRI devices can be lighter and noisier on the acquisition side, with software and accelerators restoring diagnostic image quality at the point of care. The main evidence for accelerator benefit is still imported from conventional MRI and non-MRI systems, which is the load-bearing assumption to watch.

What carries the argument

Two mechanisms carry the argument. The first is algorithmic-to-hardware mapping: SENSE's per-pixel matrix inversions, GRAPPA's k-space convolutions, and CNN inference all expose data-level parallelism that FPGA pipelines, GPU tensor cores, and ASIC systolic arrays can exploit; the review treats this structural match as why the accelerator speedups it cites are transferable. The second is the proposed evidence ladder—analytic/phantom validation, retrospective multi-center testing, prospective reader and non-inferiority trials—which is the organizational machinery meant to make AI-based low-field reconstruction reproducible and regulator-ready.

What would settle it

Measure end-to-end reconstruction time and power draw for SENSE, GRAPPA, and a CNN-based reconstruction on an edge FPGA or GPU mounted in a 0.055–0.08 T portable scanner, with the same coil array and undersampling used clinically; if the edge hardware cannot reconstruct faster than acquisition while staying within the scanner's battery budget, the claim that hardware acceleration is essential for pMRI viability is not supported.

Watch

Extended reading notes

Core claim

Central claim: hardware acceleration is essential for portable MRI viability. Edge GPUs, FPGAs, and ASICs, argues the review, are what let reconstruction and ML models run on or near the device, making real-time or near-real-time low-field imaging feasible. It pairs pMRI prototypes—an 0.08 T brain scanner with GPU-accelerated model-based reconstruction and a 0.055 T scanner with GPU-run DL super-resolution—with accelerator studies of SENSE, GRAPPA, GANs, and NUFFT from conventional MRI and non-MRI systems. It concludes FPGAs best balance latency, power, and reconfigurability, GPUs dominate prototyping, ASICs suit fixed high-volume pipelines, and edge processing avoids cloud dependence. It fu

Load-bearing premise

The paper assumes that the large speedups and low power figures measured for FPGAs, GPUs, and ASICs on conventional MRI and non-MRI workloads will transfer to low-field portable MRI systems, which have different noise statistics, coil geometries, power budgets, and regulatory constraints.

Editorial extensions

If this is right

  • Portable scanners can be built lighter and with noisier magnets, because GPU/FPGA/ASIC-based reconstruction and AI models can compensate for hardware-imposed SNR loss.
  • At the point of care, edge acceleration gives reliable real-time or near-real-time reconstruction without depending on cloud connectivity, which matters in remote or emergency settings.
  • Future pMRI systems will likely be heterogeneous: FPGA or ASIC blocks for deterministic tasks like shimming and gradient control, with GPUs handling heavier AI reconstruction.
  • Public low-field datasets and standard benchmarks become a prerequisite: AI models trained only on high-field data will not generalize, and the proposed evidence ladder is the route to regulatory acceptance.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: if the cited speedups transfer to low-field scanners, the practical bottleneck shifts from reconstruction compute to coil geometry and SNR, meaning accelerator choice alone will not determine diagnostic quality.
  • Editorial inference: the paper's essentiality claim could be settled by an on-scanner benchmark—identical SENSE, GRAPPA, and DL workloads on edge FPGA, edge GPU, and cloud GPU for the same 0.055–0.08 T system, measuring end-to-end latency and battery power.
  • Editorial inference: a consortium dataset that records coil configuration, B0, k-space trajectory, and SNR per scan—the paper's proposed metadata—would also enable transfer-learning studies and cross-device generalization tests, not just benchmarking.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The manuscript is a review of hardware acceleration in portable MRI (pMRI). It surveys pMRI system components, reconstruction algorithms (Fourier, compressed sensing, partial Fourier, SENSE, GRAPPA, deep-learning methods), and hardware accelerators (GPUs, FPGAs, ASICs). It then compares four pMRI systems—a portable brain scanner, a fast 55-mT brain MRI, an open extremity system, and the FDA-cleared Hyperfine Swoop—in Table III. The paper proposes a Low-Field MRI Consortium, an evidence ladder for regulatory validation, and discusses future directions. The central conclusion, stated in the abstract and Section VIII, is that hardware acceleration is essential for pMRI viability, that edge GPUs/FPGAs/ASICs enable real-time or near-real-time workflows, and that hardware acceleration can enhance image quality, reduce power consumption, and increase portability.

Significance. If fully supported, the review would fill a genuine gap: organized, accessible information on how hardware acceleration could be deployed in pMRI, together with a concrete plan for dataset creation and regulatory evidence. The proposed consortium and evidence ladder are constructive and potentially useful to the community. The manuscript is also useful as a compact survey of non-pMRI accelerator results (FPGA SENSE/GRAPPA, photoacoustic reconstruction, FlexiGAN) that may inform future pMRI work. However, the paper's central claim goes beyond the evidence presented. The only FDA-cleared pMRI system uses cloud GPUs rather than onboard accelerators, and the quantitative speedup evidence for FPGAs and ASICs comes predominantly from conventional MRI or non-MRI contexts. The review would be more accurate and more valuable if the conclusion were reframed as a research opportunity rather than an established necessity.

major comments (4)
  1. [Section VIII] The first bullet of Section VIII asserts that 'Hardware acceleration is essential for pMRI viability.' This is the paper's load-bearing conclusion, but it is not supported by the surveyed evidence. Section V.D reports that the only FDA-cleared pMRI (Hyperfine Swoop) relies on cloud-based GPUs and 'does not appear to rely on onboard hardware accelerators,' and Section V.B describes a 55-mT system using an NVIDIA V100, a data-center GPU, not an edge device. Nothing in the review demonstrates that pMRI without such acceleration is non-viable, or that on-/near-device acceleration is required. The conclusion should be softened to 'potentially beneficial under specific conditions' unless direct evidence is added.
  2. [Sections III.D and IV.B] The quantitative speedup evidence for FPGAs is imported from systems that are not pMRI: [15] and [16] are conventional MRI, [79] is photoacoustic tomography, and [68] is FlexiGAN outside MRI. The review assumes these speedups transfer to low-field pMRI workloads with different noise statistics, coil geometries, power budgets, and regulatory constraints. This transfer is not automatic. The manuscript should either add a concrete analysis of why the cited kernels are representative of pMRI workloads, or downgrade these numbers from evidence for pMRI to evidence for algorithmic hardware affinity.
  3. [Section IV.D / Table II] Table II and Section IV.D conclude that FPGAs are 'Excellent' and 'the most viable' for pMRI, but the table entries are derived from the non-pMRI studies discussed above. Since no FPGA/ASIC pMRI reconstruction system is reviewed in Table III, the comparative claim is not grounded in pMRI-specific measurements. Recommend adding an explicit 'evidence status' column or footnote distinguishing pMRI-specific, conventional-MRI, and non-MRI results.
  4. [Abstract and Table III] The abstract claims that 'hardware acceleration can enhance image quality, reduce power consumption, and increase portability.' The only pMRI results with acceleration are the V100-based 55-mT system (which shows reconstruction latency but no power/portability gain) and Swoop's cloud GPUs (which increase network dependence). No surveyed pMRI study shows a measured reduction in power consumption or an increase in portability due to acceleration. These outcome claims need support or should be framed as open questions.
minor comments (6)
  1. [Table III] The T1W spatial resolution for the Fast 3D brain MRI is listed as '15×1.5×1.5 mm³'. This is almost certainly a typo for '1.5×1.5×1.5 mm³'.
  2. [Section III.B] The text states 'a prior study showed that CS can recover most relevant features of an image even under 10-fold undersampling [70]' and refers to 'Figure 2'; the correct reference appears to be Figure 3, which is the compressed-sensing figure.
  3. [Section V.A] The sentence 'Among the pMRI applications reviewed in Table II...' should refer to Table III, which contains the pMRI system comparison; Table II compares accelerators.
  4. [Figure 5 caption] The caption ends with 'integrating square and. square root calculations'; the punctuation and wording need correction.
  5. [Corresponding author line] The corresponding author name appears as 'Mahmoud Meriobut' in the author block but as 'Mahmoud Meribout' in the author list; please correct the spelling.
  6. [Section II.C] The paper defines pMRI as operating at B0 less than 0.1 T, but the Open pMRI system in Table III has B0 = 295 mT. The classification of this system as 'portable MRI' should be reconciled with the stated field-strength definition.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a literature review whose conclusions are interpretive summaries of external evidence; no derivation or fitted quantity is reused as a prediction.

full rationale

This is a review/survey paper, not a derivation or modeling paper. Its central claim—that hardware acceleration is essential for pMRI viability—is presented as a synthesis of cited external results (e.g., FPGA SENSE speedups [15], FPGA GRAPPA speedups [16], GPU-accelerated 0.055-T DL reconstruction [75], FlexiGAN [68]) rather than as a quantity derived from its own definitions or fitted parameters. No equation in the paper is used both as an input and as an output; no parameter is fitted and then renamed as a prediction; and no uniqueness or ansatz is imported from the authors' own prior work. The only self-citation, [11] by Seghier and Maalej, is used in the introduction to motivate pMRI use in rescue/resource-limited settings and does not support the load-bearing acceleration claim. The paper itself explicitly flags the main limitation of its argument: Section V.D states that the FDA-cleared Hyperfine Swoop 'primarily utilizes cloud-based GPUs' and that 'the reconstruction process itself does not appear to rely on onboard hardware accelerators,' and Section VII acknowledges that 'the Swoop pMR likely relies on cloud-based GPUs... which introduces dependencies on network connectivity.' This is a genuine evidence gap—the claimed necessity of edge/FPGA/ASIC acceleration is not established by the cited examples, many of which come from conventional MRI or non-MRI domains—but it is a weakness in support, not circularity. The conclusion that hardware acceleration is 'essential' is an interpretive leap beyond the surveyed evidence, not a self-referential derivation. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

No numerical parameters are fitted; the paper is a literature review. The main burdens are the transferability of non-pMRI acceleration results and the informality of the literature selection.

assumptions (3)
  • domain assumption Low-field pMRI suffers from lower SNR and requires computational reconstruction to reach clinical utility.
    Stated in the introduction and used to motivate hardware acceleration; based on prior literature, not demonstrated in this review.
  • ad hoc to paper Speedup and power-efficiency results from non-pMRI implementations (conventional MRI, photoacoustic tomography, generic GAN accelerators) transfer to pMRI workloads.
    Used in Sections IV and V to argue for accelerator benefits in pMRI; no pMRI-specific benchmark is provided to establish this transfer.
  • domain assumption The selected literature is representative of the pMRI hardware-acceleration landscape.
    The search strategy in Section I is informal; no PRISMA-style protocol or inclusion/exclusion list is given.
invented entities (1)
  • Low-Field MRI Consortium
    purpose: Proposed multi-institutional body to curate, standardize, and openly share low-field MRI datasets and benchmarks, and to provide regulator-ready testbeds.
    The consortium is proposed in Section VII but does not yet exist and has no independent evidence of funding or operation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hardware Acceleration in Portable MRIs: State of the Art and Future Prospects." pith.science (2026). https://pith.science/paper/VV67NKHW

@misc{pith2026250906365,
  author       = {Pith},
  title        = {Pith review of: Hardware Acceleration in Portable MRIs: State of the Art and Future Prospects},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VV67NKHW}},
  note         = {Machine review of arXiv:2509.06365}
}
read the original abstract

There is a growing interest in portable MRI (pMRI) systems for point-of-care imaging, particularly in remote or resource-constrained environments. However, the computational complexity of pMRI, especially in image reconstruction and machine learning (ML) algorithms for enhanced imaging, presents significant challenges. Such challenges can be potentially addressed by harnessing hardware application solutions, though there is little focus in the current pMRI literature on hardware acceleration. This paper bridges that gap by reviewing recent developments in pMRI, emphasizing the role and impact of hardware acceleration to speed up image acquisition and reconstruction. Key technologies such as Graphics Processing Units (GPUs), Field-Programmable Gate Arrays (FPGAs), and Application-Specific Integrated Circuits (ASICs) offer excellent performance in terms of reconstruction speed and power consumption. This review also highlights the promise of AI-powered reconstruction, open low-field pMRI datasets, and innovative edge-based hardware solutions for the future of pMRI technology. Overall, hardware acceleration can enhance image quality, reduce power consumption, and increase portability for next-generation pMRI technology. To accelerate reproducible AI for portable MRI, we propose forming a Low-Field MRI Consortium and an evidence ladder (analytic/phantom validation, retrospective multi-center testing, prospective reader and non-inferiority trials) to provide standardized datasets, benchmarks, and regulator-ready testbeds.

Figures

Figures reproduced from arXiv: 2509.06365 by the authors.

Figure 1
Figure 1. illustrates the flow diagram of literature search strategy. Hundreds of papers were reviewed in Scopus and Google Scholar databases using inclusive keywords: “pMRI”, “real-time pMRI”, and “Hardware accelerator for pMRI” for the duration of 2020-2025. The keyword “MRI” was excluded since the focus was on pMRI only. Literature that uses hardware accelerators like GPU, FPGA and ASIC for functions other than accelerator… view at source ↗
Figure 9
Figure 9. FIGURE 9 [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

123 extracted references · 76 canonical work pages

  1. [15]

    FPGA-based hardware accelerator for sense (a parallel MR image reconstruction method),

    O. Inam, A. Basit, M. Qureshi, and H. Omer, “FPGA-based hardware accelerator for sense (a parallel MR image reconstruction method),” Computers in Biology and Medicine, vol. 117, p. 103598, 2020

    2020

  2. [16]

    Accelerating grappa reconstruction using soc design for real -time cardiac MRI,

    A. Basit, O. Inam, and H. Omer, “Accelerating grappa reconstruction using soc design for real -time cardiac MRI,” Computers in Biology and Medicine, vol. 160, p. 107008, 2023

    2023

  3. [79]

    FPGA acceleration of image reconstruction for real -time photoacoustic tomography,

    Z. Gao, Y. Shen, D. Jiang, F. Liu, F. Gao, and F. Gao, “FPGA acceleration of image reconstruction for real -time photoacoustic tomography,” 2022

    work page 2022

  4. [68]

    Flexigan: An end-to-end framework for FPGA acceleration of generative adversarial networks,

    B. Khaleghi, A. Shao, T. Guan, A. Tran, A. Joshi, M. Abadi, and V. J. Reddi, “Flexigan: An end-to-end framework for FPGA acceleration of generative adversarial networks,” in 26th IEEE International Symposium on Field -Programmable Custom Computing Machines (FCCM), pp. 35–42, IEEE, 2018. VOLUME XX, 2017 7

    work page 2018

  5. [1]

    Low -cost and portable MRI,

    L. L. Wald, P. C. McDaniel, T. Witzel, J. P. Stockmann, and C. Z. Cooley, “Low -cost and portable MRI,” Journal of Magnetic Resonance Imaging, vol. 52, pp. 686–696, Sep 2020

    2020

  6. [2]

    Low -field MRI: Clinical promise and challenges,

    T. C. Arnold, C. W. Freeman, B. Litt, and J. M. Stein, “Low -field MRI: Clinical promise and challenges,” Journal of Magnetic Resonance Imaging, vol. 57, no. 1, pp. 25–44, 2023

    2023

  7. [3]

    Undisclosed and undetected foreign bodies during MRI screening resulting in a potentially serious outcome,

    C. A. James, A. Karacozoff, and F. G. Shellock, “Undisclosed and undetected foreign bodies during MRI screening resulting in a potentially serious outcome,” Magnetic Resonance Imaging, vol. 31, no. 4, pp. 630–633, 2013

    2013

  8. [4]

    Portable, bedside, low -field magnetic resonance imaging for evaluation of intracerebral hemorrhage,

    K. N. Sheth, M. H. MacDougall, W. T. Kimberly, et al., “Portable, bedside, low -field magnetic resonance imaging for evaluation of intracerebral hemorrhage,” Nature Communications, vol. 12, no. 5119, 2021

    2021

Show all 123 references
  1. [5]

    The rise and efficiency of low field portable MRI scanners,

    T. Pires and J. M.M., “The rise and efficiency of low field portable MRI scanners,” Journal of Health Allied SciencesNU, June 2023. Preprint

    2023

  2. [6]

    Current role of portable MRI in diagnosis of acute neurological conditions,

    A. Shoghli, D. Chow, E. Kuoy, and V. Yaghmai, “Current role of portable MRI in diagnosis of acute neurological conditions,” Frontiers in Neurology, vol. 14, 2023

    2023

  3. [7]

    Hyperfine swoop portable MRI,

    Hyperfine, Inc., “Hyperfine swoop portable MRI,” 2024. Accessed: 2024-10-14

    2024

  4. [8]

    Qhs student helps make neurological mri testing accessible to remote communities,

    Q. H. Sciences, “Qhs student helps make neurological mri testing accessible to remote communities,” 2023. Accessed: 2024-10-14

    2023

  5. [9]

    Design and test of an open portable MRI system,

    S. Chauvi`ere, L. Belguerras, T. Lubin, and S. Mezani, “Design and test of an open portable MRI system,” COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, 2022

    2022

  6. [10]

    A wrist MRI for the International Space Station,

    G. E. Sarty, A. Scott, K. Turek, P. Liszkowski, J. Sharp, and B. Tomanek, “A wrist MRI for the International Space Station,” in 65th International Astronautical Congress (Toronto, Canada), International Astronautical Federation (IAF), 2014

    2014

  7. [11]

    Rescue equipment should include portable medical imaging systems,

    M. L. Seghier and N. Maalej, “Rescue equipment should include portable medical imaging systems,” Science Bulletin (Beijing), vol. 69, pp. 1819–1822, June 30, 2024. Epub 2024 May 3

    2024

  8. [12]

    Pushing the limits of low -cost ultra -low-field MRI by dual - acquisition deep learning 3d super resolution,

    V. Lau, L. Xiao, Y. Zhao, S. Su, Y. Ding, C. Man, X. Wang, A. Tsang, P. Cao, G. K. K. Lau, G. K. K. Leung, A. T. L. Leong, and E. X. Wu, “Pushing the limits of low -cost ultra -low-field MRI by dual - acquisition deep learning 3d super resolution,” Magnetic Re sonance in Medic...

    2023

  9. [13]

    Suitability of recent hardware accelerators (DSPs, FPGAs, andGPUs) for computer vision and image processing algorithms,

    A. HajiRassouliha, A. J. Taberner, M. P. Nash, and P. M. Nielsen, “Suitability of recent hardware accelerators (DSPs, FPGAs, andGPUs) for computer vision and image processing algorithms,” Signal Processing: Image Communication, vol. 68, pp. 101–119, 2018

    2018

  10. [14]

    A survey of GPU-based acceleration techniques in MRI reconstructions,

    H. Wang, H. Peng, Y. Chang, and D. Liang, “A survey of GPU-based acceleration techniques in MRI reconstructions,” Quantitative Imaging in Medicine and Surgery, vol. 8, no. 2, 2018

    2018

  11. [17]

    New challenges and opportunities for low-field MRI,

    E. Anoardo and G. G. Rodriguez, “New challenges and opportunities for low-field MRI,” Journal of Magnetic Resonance Open, vol. 14-15, p. 100086, 2023

    2023

  12. [18]

    Publisher correction: Brain imaging with portable low -field MRI,

    W. T. Kimberly, A. J. Sorby -Adams, A. G. Webb, et al., “Publisher correction: Brain imaging with portable low -field MRI,” Nature Reviews Bioengineering, vol. 1, p. 680, 2023

    2023

  13. [19]

    Applications, limitations and advancements of ultra -low-field magnetic resonance imaging: A scoping review,

    S. Wilson, L. K. Tan, J. Zhang, and B. J. MacIntosh, “Applications, limitations and advancements of ultra -low-field magnetic resonance imaging: A scoping review,” Magnetic Resonance Imaging, vol. 102, pp. 44–53, 2023

    2023

  14. [20]

    Prepolarized magnetic resonance imaging around metal orthopedic implants,

    R. D. Venook, N. I. Matter, M. Ramachandran, S. E. Ungersma, G. E. Gold, N. J. Giori, A. Macovski, G. C. Scott, and S. M. Conolly, “Prepolarized magnetic resonance imaging around metal orthopedic implants,” Magnetic Resonance in Medicine, vol. 56, no. 1, p p. 177– 186, 2006

    2006

  15. [21]

    Basic principles of magnetic resonance imaging,

    R.-J. M. van Geuns, P. A. Wielopolski, H. G. de Bruin, B. J. Rensing, P. M. van Ooijen, M. Hulshoff, M. Oudkerk, and P. J. de Feyter, “Basic principles of magnetic resonance imaging,” Progress in Cardiovascular Diseases, vol. 42, no. 2, pp. 149–156, 1999. Advances in Coronary Imaging

    1999

  16. [22]

    Clinician’s guide to the basic principles of MRI,

    S. Azhar and L. R. Chong, “Clinician’s guide to the basic principles of MRI,” Postgraduate Medical Journal, vol. 99, pp. 894 –903, 09 2022

    2022

  17. [23]

    A low -cost and shielding-free ultra-low-field brain MRI scanner,

    Y. Liu, A. T. Leong, Y. Zhao, et al., “A low -cost and shielding-free ultra-low-field brain MRI scanner,” Nature Communications, vol. 12,p 7238, 2021

    2021

  18. [24]

    Permanent magnet materials and their applications p. campbell,

    W. Reitz, “Permanent magnet materials and their applications p. campbell,” Materials and Manufacturing Processes, vol. 13, no. 3, pp. 469–470, 1998

    1998

  19. [25]

    Smco magnets vs neodymium magnets: A detailed comparison,

    Stanford Magnets, “Smco magnets vs neodymium magnets: A detailed comparison,” 2024. Accessed: 2024-11-25

    2024

  20. [26]

    Samarium -cobalt magnet information,

    M. Searcher, “Samarium -cobalt magnet information,” 2024. Accessed: 2024-11-25. VOLUME XX, 2017 7

    2024

  21. [27]

    Neodymium -iron-boron magnet information,

    M. Searcher, “Neodymium -iron-boron magnet information,” 2024. Accessed: 2024-11-25

    2024

  22. [28]

    Compact low field magnetic resonance imaging magnet: Design and optimization,

    M. Sciandrone, G. Placidi, L. Testa, and A. Sotgiu, “Compact low field magnetic resonance imaging magnet: Design and optimization,” Review of scientific instruments, vol. 71, no. 3, pp. 1534–1538, 2000

    2000

  23. [29]

    Design and simulations of low cost and low magnetic field mri system,

    S. K. Ghosh, V. Thakur, and S. R. Chowdhury, “Design and simulations of low cost and low magnetic field mri system,” 2017 Eleventh International Conference on Sensing Technology (ICST), pp. 1–6, 2017

    2017

  24. [30]

    Ultra -high field NMR and MRI—the role of magnet technology to increase sensitivity and specificity,

    E. Moser, E. Laistler, F. Schmitt, and G. Kontaxis, “Ultra -high field NMR and MRI—the role of magnet technology to increase sensitivity and specificity,” Frontiers in Physics, vol. 5, p. 33, 2017

    2017

  25. [31]

    A high temperature superconducting imaging coil for low-field MRI,

    H.-S. Cheong, J. M. Wild, N. M. Alford, I. Valkov, C. P. Randell, and M. N. J. Paley, “A high temperature superconducting imaging coil for low-field MRI,” Concepts in Magnetic Resonance Part B -Magnetic Resonance Engineering, vol. 37, pp. 56–64, 2010

    2010

  26. [32]

    Shim coil design, limitations and implications,

    P. Jezzard, “Shim coil design, limitations and implications,” in Abstracts from the International Society of Magnetic Resonance in Medicine (ISMRM) Annual Meeting, (Seattle, WA, USA), May 2006. FMRIB Centre, John Radcliffe Hospital, Headington, Oxford, England

    2006

  27. [33]

    Passive shimming of MR magnets: algorithm, hardware, and results,

    B. C. P. Dorri, M. E. Vermilyea, and W. E. Toffolo, “Passive shimming of MR magnets: algorithm, hardware, and results,” IEEE Transactions on Applied Superconductivity, vol. 3, pp. 254 –257, 1993

    1993

  28. [34]

    Shimming toolbox: An opensource software toolbox for b0 and b1 shimming in MRI,

    A. D’Astous, G. Cereza, D. Papp, K. M. Gilbert, J. P. Stockmann, E. Alonso-Ortiz, and J. Cohen -Adad, “Shimming toolbox: An opensource software toolbox for b0 and b1 shimming in MRI,” Magnetic Resonance in Medicine, vol. 89, pp. 1401 – 1417, 2022

    2022

  29. [35]

    Theory of gradient coil design methods for magnetic resonance imaging,

    S. Hidalgo -Tobon, “Theory of gradient coil design methods for magnetic resonance imaging,” Concepts in Magnetic Resonance Part A, vol. 36A, no. 4, pp. 223–242, 2010

    2010

  30. [36]

    Geometric distortion correction in gradient-echo imaging by use of dynamic time warping,

    S. A. Kannengießer, Y. Wang, and E. M. Haacke, “Geometric distortion correction in gradient-echo imaging by use of dynamic time warping,” Magnetic Resonance in Medicine, vol. 42, no. 3, pp. 585 – 590, 1999

    1999

  31. [37]

    Image distortion correction for mri in low field permanent magnet systems with strong B0 inhomogeneity and gradient field nonlinearities,

    K. Koolstra, T. O’Reilly, P. B¨ornert, and A. Webb, “Image distortion correction for mri in low field permanent magnet systems with strong B0 inhomogeneity and gradient field nonlinearities,” Magnetic Resonance Materials in Physics, Biology and Medicine, vol. 34, no. 5, pp. 63...

    2021

  32. [38]

    Distortion -corrected image reconstruction with deep learning on an MRI-linac,

    S. Shan, Y. Gao, P. Z. Y. Liu, B. Whelan, H. Sun, B. Dong, F. Liu, and D. E. J. Waddington, “Distortion -corrected image reconstruction with deep learning on an MRI-linac,” 2023

    2023

  33. [39]

    A programmable eddy -current compensation system for mri and localized spectroscopy,

    M. Fry, S. Pittard, I. Summers, W. Vennart, and F. Goldie, “A programmable eddy -current compensation system for mri and localized spectroscopy,” Journal of magnetic resonance imaging: JMRI, vol. 7, pp. 455–8, 03 1997

    1997

  34. [40]

    Rf coils: A practical guide for nonphysicists,

    B. Gruber, M. Froeling, T. Leiner, and D. W. Klomp, “Rf coils: A practical guide for nonphysicists,” Journal of Magnetic Resonance Imaging, vol. 48, pp. 590–604, 2018

    2018

  35. [41]

    Design and implementation of split -leg type elliptical whole -body birdcage RF coil at 1.5 T MRI,

    S. Kumar, H.-J. Chung, Y.-J. Jeong, H.-K. Lee, and C.-H. Oh, “Design and implementation of split -leg type elliptical whole -body birdcage RF coil at 1.5 T MRI,” Applied Sciences, vol. 11, p. 7448, 2021

    2021

  36. [42]

    Basic principles of and practical guide to clinical MRI radiofrequency coils,

    W. E. Kwok, “Basic principles of and practical guide to clinical MRI radiofrequency coils,” Radiographics, vol. 42, pp. 898–918, 2022

    2022

  37. [43]

    W. E. Kwok, “Basic principles of and practical guide to clinical MRI

  38. [44]

    radiofrequency coils,” RadioGraphics, vol. 42, no. 3, pp. 898 –918, 2022

    2022

  39. [45]

    Recent advances in parallel imaging for MRI,

    [44] J. Hamilton, D. Franson, and N. Seiberlich, “Recent advances in parallel imaging for MRI,” Progress in Nuclear Magnetic Resonance Spectroscopy, vol. 101, pp. 71–95, 2017

    2017

  40. [46]

    An introduction to coil array design for parallel MRI,

    M. A. Ohliger and D. K. Sodickson, “An introduction to coil array design for parallel MRI,” NMR in Biomedicine, vol. 19, no. 3, pp. 300– 315, 2006

    2006

  41. [47]

    Us regulatory considerations for low field magnetic resonance imaging systems,

    D. M. Krainak, R. Zeng, N. Li, T. O. Woods, and J. G. Delfino, “Us regulatory considerations for low field magnetic resonance imaging systems,” MAGMA, vol. 36, no. 3, pp. 347–354, 2023

    2023

  42. [48]

    MRI (Magnetic Resonance Imaging) - Information for Industry,

    U.S. Food and Drug Administration, “MRI (Magnetic Resonance Imaging) - Information for Industry,” 2024. Accessed: 2024-10-28

    2024

  43. [49]

    Geneva, Switzerland: International Electrotechnical Commission, 4.2 ed., 2022

    International Electrotechnical Commission, IEC 60601-2-33: Medical Electrical Equipment - Part 2 -33: Particular Requirements for the Basic Safety and Essential Performance of Magnetic Resonance Equipment for Medical Diagnosis. Geneva, Switzerland: International Electrotechnic...

    2022

  44. [50]

    Rosslyn, VA: National Electrical Manufacturers Association, 2008

    National Electrical Manufacturers Association, NEMA MS 1 -2008: Determination of Signal-to-Noise Ratio (SNR) and Image Uniformity for Magnetic Resonance Imaging (MRI) Systems. Rosslyn, VA: National Electrical Manufacturers Association, 2008

    2008

  45. [51]

    Rosslyn, VA: National Electrical Manufacturers Association, 2023

    National Electrical Manufacturers Association, NEMA MS 4 -2023: Determination of Signal -to-Noise Ratio (SNR) in Diagnostic Magnetic Resonance Imaging. Rosslyn, VA: National Electrical Manufacturers Association, 2023

    2023

  46. [52]

    Recent advances in parallel imaging for mri,

    J. Hamilton, D. Franson, and N. Seiberlich, “Recent advances in parallel imaging for mri,” Progress in Nuclear Magnetic Resonance Spectroscopy, vol. 101, pp. 71–95, 2017

    2017

  47. [53]

    Sense: Sensitivity encoding for fast MRI,

    K. P. Pruessmann, M. Weiger, M. B. Scheidegger, and P. Boesiger, “Sense: Sensitivity encoding for fast MRI,” Magnetic Resonance in Medicine, vol. 42, no. 5, pp. 952–962, 1999

    1999

  48. [54]

    Qrdecomposition based sense reconstruction using parallel architecture,

    I. Ullah, H. Nisar, H. Raza, M. Qasim, O. Inam, and H. Omer, “Qrdecomposition based sense reconstruction using parallel architecture,” Computers in Biology and Medicine, vol. 95, pp. 1–12, 2018

    2018

  49. [55]

    A 2d -grappa algorithm with a boomerang kernel for 3d mri data accelerated along two phaseencoding directions,

    S. Shin, Y. Han, and J. -Y. Chung, “A 2d -grappa algorithm with a boomerang kernel for 3d mri data accelerated along two phaseencoding directions,” Sensors, vol. 23, no. 1, 2023. [55] M. S. Hansen and T. S. Sørensen, “Gadgetron: An open -source framework for medical image reco...

    2023

  50. [56]

    Deep learning for accelerated and robust mri reconstruction: A review,

    R. Heckel, M. Jacob, A. Chaudhari, O. Perlman, and E. Shimron, “Deep learning for accelerated and robust mri reconstruction: A review,” arXiv preprint, 2024

    2024

  51. [57]

    MRI image analysis methods and applications: an algorithmic perspective using brain tumors as an exemplar,

    V. Vadmal, G. Junno, C. Badve, W. Huang, K. A. Waite, and J. S. Barnholtz-Sloan, “MRI image analysis methods and applications: an algorithmic perspective using brain tumors as an exemplar,” Neuro - Oncology Advances, vol. 2, p. vdaa049, 04 2020

    2020

  52. [58]

    Fast MRI reconstruction using deep learning-based compressed sensing: A systematic review,

    M. Safari, Z. Eidex, C. -W. Chang, R. L. J. Qiu, and X. Yang, “Fast MRI reconstruction using deep learning-based compressed sensing: A systematic review,” arXiv preprint, 2024

    2024

  53. [59]

    Generative adversarial networks (GAN) powered fast magnetic resonance imaging —mini review, comparison and perspectives,

    G. Yang, J. Lv, Y. Chen, J. Huang, and J. Zhu, “Generative adversarial networks (GAN) powered fast magnetic resonance imaging —mini review, comparison and perspectives,” arXiv preprint, 2021

    2021

  54. [60]

    Image quality assessment for magnetic resonance imaging,

    S. Kastryulin, J. Zakirov, N. Pezzotti, and D. V. Dylov, “Image quality assessment for magnetic resonance imaging,” arXiv preprint, 2022

    2022

  55. [61]

    Deep learning for undersampled MRI reconstruction,

    C. M. Hyun, H. P. Kim, S. M. Lee, S. M. Lee, and J. Seo, “Deep learning for undersampled MRI reconstruction,” Physics in Medicine & Biology, vol. 63, 2017

    2017

  56. [62]

    A CNN accelerator on FPGA using depthwise separable convolution,

    L. Luo, F. Zhang, S. Zhang, W. Liu, and X. Zhao, “A CNN accelerator on FPGA using depthwise separable convolution,” arXiv preprint arXiv:1809.01536, 2018

    2018 arXiv

  57. [63]

    A comparative study of cnn -based super -resolution methods in mri reconstruction,

    W. Zeng, J. Peng, S. Wang, Z. Li, Q. Liu, and D. Liang, “A comparative study of cnn -based super -resolution methods in mri reconstruction,” in 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), pp. 1678–1682, IEEE, 2019

    2019

  58. [64]

    Generative adversarial networks (GAN) powered fast magnetic resonance imaging – mini review, comparison and perspectives,

    G. Yang, J. Lv, Y. Chen, J. Huang, and J. Zhu, “Generative adversarial networks (GAN) powered fast magnetic resonance imaging – mini review, comparison and perspectives,” 2021

    2021

  59. [65]

    The role of generative adversarial networks in brain MRI: a scoping review,

    H. Ali, M. R. Biswas, F. Mohsen, U. Shah, A. Alamgir, O. Mousa, and Z. Shah, “The role of generative adversarial networks in brain MRI: a scoping review,” Insights into Imaging, vol. 13, 2022

    2022

  60. [66]

    Super -resolution of magnetic resonance images using generative adversarial networks,

    J. Guerreiro, P. Tom´as, N. Garcia, and H. Aidos, “Super -resolution of magnetic resonance images using generative adversarial networks,” Computerized Medical Imaging and Graphics, vol. 108, p. 102280, 2023

    2023

  61. [67]

    Soup-gan: Super-resolution MRI using generative adversarial networks,

    K. Zhang, H. Hu, K. Philbrick, G. M. Conte, J. D. Sobek, P. Rouzrokh, and B. J. Erickson, “Soup-gan: Super-resolution MRI using generative adversarial networks,” arXiv preprint arXiv:2106.02599, 2021

    2021 arXiv

  62. [69]

    Compressed sensing MRI: a review of the clinical literature,

    O. N. Jaspan, R. Fleysher, and M. L. Lipton, “Compressed sensing MRI: a review of the clinical literature,” British Journal of Radiology, vol. 88, no. 1056, p. 20150487, 2015. Epub 2015 Sep 24

    2015

  63. [70]

    Compressed sensing MRI,

    M. Lustig, D. L. Donoho, J. M. Santos, and J. M. Pauly, “Compressed sensing MRI,” IEEE Signal Processing Magazine, vol. 25, no. 2, pp. 72– 82, 2008

    2008

  64. [71]

    Quantitative evaluation of several partial Fourier reconstruction algorithms used in MRI,

    G. McGibney, M. R. Smith, S. T. Nichols, and A. Crawley, “Quantitative evaluation of several partial Fourier reconstruction algorithms used in MRI,” Magnetic Resonance in Medicine, vol. 30, no. 1, pp. 51–59, 1993

    1993

  65. [72]

    Partial fourier partially parallel imaging,

    M. Bydder and M. D. Robson, “Partial fourier partially parallel imaging,” Magnetic Resonance in Medicine, vol. 53, no. 6, pp. 1393– 1401, 2005

    2005

  66. [73]

    Hardware architectures for real-time medical imaging,

    E. Alca´ın, P. R. Fern´andez, R. Nieto, A. S. Montemayor, J. Vilas, A. Galiana-Bordera, P. M. Mart´ınez -Giron´es, C. P. de -la Lastra, B. Rodr´ıguez-Vila, M. Bonet, C. Rodr´ıguez-S´anchez, I. Yahyaoui, N. Malpica, S. Borromeo, F. Machado, and A. Torrado -Carvajal, “Hardware a...

    2021

  67. [74]

    Hardware architectures for real-time medical imaging,

    E. Alca´ın, P. R. Fern´andez, R. Nieto, A. S. Montemayor, J. Vilas, A. Galiana-Bordera, P. M. Martinez -Girones, C. Prieto -de-la Lastra, B. Rodriguez-Vila, M. Bonet, C. Rodriguez -Sanchez, I. Yahyaoui, N. Malpica, S. Borromeo, F. Machado, and A. Torrado -Carvajal, “Hardware a...

    2021

  68. [75]

    Deep learning enabled fast 3d brain MRI at 0.055 Tesla,

    C. Man, V. Lau, S. Su, Y. Zhao, L. Xiao, Y. Ding, G. K. K. Leung, A. T. L. Leong, and E. X. Wu, “Deep learning enabled fast 3d brain MRI at 0.055 Tesla,” Science Advances, vol. 9, no. 38, p. eadi9327, 2023

    2023

  69. [76]

    ASIC modelling of sense for parallel MRI,

    S. A. Qazi, M. F. Siddiqui, J. Jacob Wikner, and H. Omer, “ASIC modelling of sense for parallel MRI,” Computers in Biology and Medicine, vol. 109, pp. 53–61, 2019

    2019

  70. [77]

    Accelerating advanced MRI reconstructions on GPUs,

    S. Stone, J. Haldar, S. Tsao, W. m.W. Hwu, B. Sutton, and Z.-P. Liang, “Accelerating advanced MRI reconstructions on GPUs,” Journal of Parallel and Distributed Computing, vol. 68, no. 10, pp. 1307 –1318,

  71. [78]

    GPU-accelerated dynamic functional connectivity analysis for functional MRI data,

    D. Akgun ¨n, U. Sakog˘lu, J. Esquivel, B. Adinoff, and M. Mete, “GPU-accelerated dynamic functional connectivity analysis for functional MRI data,” Computerized Medical Imaging and Graphics, vol. 43, pp. 53– 63, 2015

    2015

  72. [80]

    Parallel 2d fft implementation on FPGA suitable for real -time mr image processing,

    L. Li and A. M. Wyrwicz, “Parallel 2d fft implementation on FPGA suitable for real -time mr image processing,” Review of Scientific Instruments, vol. 89, p. 093706, 09 2018

    2018

  73. [81]

    FPGA implementation of real -time sense reconstruction using pre - scan and emaps sensitivities,

    M. F. Siddiqui, A. W. Reza, A. Shafique, H. Omer, and J. Kanesan, “FPGA implementation of real -time sense reconstruction using pre - scan and emaps sensitivities,” Magnetic Resonance Imaging, vol. 44, pp. 82–91, 2017

    2017

  74. [82]

    FPGA implementation of a real -time parallel MRI reconstruction,

    M. F. Siddiqui and H. Omer, “FPGA implementation of a real -time parallel MRI reconstruction,” December 2017

    2017

  75. [83]

    Dlau: A scalable deep learning accelerator unit on fpga,

    C. Wang, L. Gong, Q. Yu, X. Li, Y. Xie, and X. Zhou, “Dlau: A scalable deep learning accelerator unit on fpga,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 36, no. 3, pp. 513–517, 2017

    2017

  76. [84]

    Google supercharges machine learning tasks with custom chip,

    Google Cloud AI Team, “Google supercharges machine learning tasks with custom chip,” 2023. Accessed: June 18, 2024

    2023

  77. [85]

    In -datacenter performance analysis of a tensor processing unit,

    C. Y. Norman, P. Jouppi, and N. P. et.al, “In -datacenter performance analysis of a tensor processing unit,” 2017

    2017

  78. [86]

    Jigsaw: A slice-and-dice approach to non -uniform FFT acceleration for MRI image reconstruction,

    B. L. West, J. A. Fessler, and T. F. Wenisch, “Jigsaw: A slice-and-dice approach to non -uniform FFT acceleration for MRI image reconstruction,” in 2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS), pp. 821–831, IEEE, 2021

    2021

  79. [87]

    Low power and complexity implementation of the modified FFT with a new bit-slicing scheme,

    S. Qadeer, H. Keerthan, A. Syed, and M. Khan, “Low power and complexity implementation of the modified FFT with a new bit-slicing scheme,” Journal of The Institution of Engineers (India): Series B, vol. 104, 10 2023

    2023

  80. [88]

    Unveiling the powerhouses of AI: A comprehensive study of GPU, FPGA, and ASIC accelerators,

    Y. Shi, “Unveiling the powerhouses of AI: A comprehensive study of GPU, FPGA, and ASIC accelerators,” Applied and Computational Engineering, vol. 50, pp. 97–105, 03 2024

    2024

  81. [89]

    A portable scanner for brain MRI,

    C. Cooley, P. C. McDaniel, J. Stockmann, S. Srinivas, S. Cauley, M. ´ Sliwiak, C. R. Sappo, C. F. Vaughn, B. Gu´erin, M. Rosen, M. Lev, and L. Wald, “A portable scanner for brain MRI,” Nature biomedical engineering, vol. 5, pp. 229 – 239, 2020

    2020

  82. [90]

    Theory of gradient coil design methods for magnetic resonance imaging,

    S. Hidalgo -Tob´on, “Theory of gradient coil design methods for magnetic resonance imaging,” Concepts in Magnetic Resonance Part A, vol. 36, pp. 223–242, 2010

    2010

  83. [91]

    MRI using radiofrequency magnetic field phase gradients,

    J. C. Sharp and S. B. King, “MRI using radiofrequency magnetic field phase gradients,” Magnetic Resonance in Medicine, vol. 63, 2010

    2010

  84. [92]

    Model-based image reconstruction for MRI,

    [92] J. A. Fessler, “Model-based image reconstruction for MRI,” IEEE Signal Processing Magazine, vol. 27, no. 4, pp. 81–89, 2010

    2010

  85. [93]

    The wu -minn human connectome project: An overview,

    D. C. Van Essen, S. M. Smith, D. M. Barch, T. E. Behrens, E. Yacoub, and K. Ugurbil, “The wu -minn human connectome project: An overview,” NeuroImage, vol. 80, pp. 62 –79, 2013. Mapping the Connectome

    2013

  86. [94]

    Hyperfine, inc. dl image reconstruction white paper

    M. Sofka, K. Siddiqui, and E. Siegel, “Hyperfine, inc. dl image reconstruction white paper.” White Paper, Hyperfine, Inc., 2021. LBL- 001364 v1

    2021

  87. [95]

    Ultra-low-field paediatric MRI in low- and middle -income countries: Super -resolution using a multiorientation u -net,

    L. Baljer, Y. Zhang, N. J. Bourke, K. A. Donald, L. E. Bradford, J. E. Ringshaw, S. R. Williams, S. C. L. Deoni, S. C. R. Williams, K. S. S. Team, F. V´aˇsa, and R. J. Moran, “Ultra-low-field paediatric MRI in low- and middle -income countries: Super -resolution using a multio...

    2025

  88. [96]

    Dagan: Deep dealiasing generative adversarial networks for fast compressed sensing mri reconstruction,

    G. Yang, S. Yu, H. Dong, G. Slabaugh, P. L. Dragotti, X. Ye, F. Liu, S. Arridge, J. Keegan, Y. Guo, and D. Firmin, “Dagan: Deep dealiasing generative adversarial networks for fast compressed sensing mri reconstruction,” IEEE Transactions on Medical Imaging, vol. 37, no. 6, pp....

    2018

  89. [97]

    Accelerating low -field MRI: Compressed sensing and AI for fast noiserobust imaging,

    E. Shimron, S. Shan, J. Grover, N. Koonjoo, S. Shen, T. Boele, A. J. Sorby-Adams, J. E. Kirsch, M. S. Rosen, and D. E. Waddington, “Accelerating low -field MRI: Compressed sensing and AI for fast noiserobust imaging,” arXiv preprint arXiv:2411.06704, 2024

    2024

  90. [98]

    Dual -domain self-supervised learning for accelerated non -cartesian MRI reconstruction,

    B. Zhou, J. Schlemper, N. Dey, S. S. Mohseni Salehi, K. Sheth, C. Liu, J. S. Duncan, and M. Sofka, “Dual -domain self-supervised learning for accelerated non -cartesian MRI reconstruction,” Medical Image Analysis, vol. 81, p. 102538, 2022

    2022

  91. [99]

    Two -dimensional imaging in a lightweight portable MRI scanner without gradient coils,

    C. Z. Cooley, J. P. Stockmann, B. D. Armstrong, M. Sarracanie, M. H. Lev, M. S. Rosen, and L. L. Wald, “Two -dimensional imaging in a lightweight portable MRI scanner without gradient coils,” Magnetic Resonance in Medicine, vol. 73, no. 2, pp. 872–883, 2015

    2015

  92. [100]

    Unity: A low -field magnetic resonance neuroimaging initiative to characterize neurodevelopment in low and middle -income settings,

    e. F. Abate, A. Adu -Amankwah, “Unity: A low -field magnetic resonance neuroimaging initiative to characterize neurodevelopment in low and middle -income settings,” Developmental Cognitive Neuroscience, vol. 69, p. 101397, 2024

    2024

  93. [101]

    Denoising simulated low -field MRI (70mT) using denoising autoencoders (DAE) and cycle - consistent generative adversarial networks (cycle -GAN)

    Vega F, Addeh A, MacDonald ME. “Denoising simulated low -field MRI (70mT) using denoising autoencoders (DAE) and cycle - consistent generative adversarial networks (cycle -GAN)”. arXiv preprint arXiv:2307.06338. 2023 Jul 12

    2023 arXiv

  94. [102]

    Ali, H., Biswas, M.R., Mohsen, F. et al. The role of generative adversarial networks in brain MRI: a scoping review. Insights Imaging 13, 98 (2022). https://doi.org/10.1186/s13244-022-01237-0

    2022 doi

  95. [103]

    Makhlouf, A., Maayah, M., Abughanam, N. et al. The use of generative adversarial networks in medical image augmentation. Neural Comput & Applic 35, 24055 –24068 (2023). https://doi.org/10.1007/s00521-023-09100-z

    2023 doi

  96. [104]

    MRI at low field: A review of software solutions for improving SNR

    Ayde R, Vornehm M, Zhao Y, Knoll F, Wu EX, Sarracanie M. MRI at low field: A review of software solutions for improving SNR. NMR Biomed. 2025 Jan;38(1):e5268. doi: 10.1002/nbm.5268. Epub 2024 Oct 7. PMID: 39375036; PMCID: PMC11605168

    2025 doi

  97. [105]

    Aug 13, 2020

    Hyperfine, “ Hyperfine Receives 510(k) Clearance from US FDA . Aug 13, 2020

    2020

  98. [106]

    Premarket Approval (PMA)

    U.S. Food and Drug Administration, “Premarket Approval (PMA)”

  99. [107]

    IEC 62304

    International Electrotechnical Commission, “IEC 62304”

  100. [108]

    ISO 14971:2019

    ISO, “ISO 14971:2019”, edition 3, 2019

    2019

  101. [109]

    ISO 13485, Medical devices

    ISO, “ISO 13485, Medical devices”

  102. [110]

    Lustig, et al., Kt SPARSE: high frame rate dynamic MRI exploiting spatiotemporal sparsity, in: Proceedings of the 13th Annual Meeting of ISMRM 2420, 2006

    M. Lustig, et al., Kt SPARSE: high frame rate dynamic MRI exploiting spatiotemporal sparsity, in: Proceedings of the 13th Annual Meeting of ISMRM 2420, 2006. Seattle

    2006

  103. [111]

    Usman, et al., k-t group sparse: a method for accelerating dynamic MRI, Magn

    M. Usman, et al., k-t group sparse: a method for accelerating dynamic MRI, Magn. Reson. Med. 66 (4) (2011) 1163–1176. VOLUME XX, 2017 7

    2011

  104. [112]

    Classification of Parkinson’s Disease in Patch -Based MRI of Substantia Nigra

    Hussain, S.S.; Degang, X.; Shah, P.M.; Islam, S.U.; Alam, M.; Khan, I.A.; Awwad, F.A.; Ismail, E.A.A. Classification of Parkinson’s Disease in Patch -Based MRI of Substantia Nigra. Diagnostics 2023, 13, 2827. https://doi.org/10.3390/diagnostics13172827

    2023 doi

  105. [113]

    L. Su; K. A. Alattas; W. Ayadi; Z. Lynton; E. Ghaderpour; A. Mohammadzadeh, C. Zhang. A novel fuzzy reinforcement learning approach for personalization deep brain stimulation with communication delay. Biomedical Signal Processing and Control 2025, 106, 107736. https://doi.org/...

    2025

  106. [114]

    NVIDIA T4 Tensor Core GPU

    NVIDIA, “ NVIDIA T4 Tensor Core GPU”

  107. [115]

    Hyperfine partners with NVIDIA to enhance portable MRI tech

    “Hyperfine partners with NVIDIA to enhance portable MRI tech”, URL: Investing.com. Article published on 03/17/2025

    2025

  108. [116]

    Xilinx Adaptive Compute Acceleration Platform: Versal Architecture,

    B. Gaide et al., “Xilinx Adaptive Compute Acceleration Platform: Versal Architecture,” in International Symposium on Field - Programmable Gate Arrays (FPGA), 2019, pp. 84–93

    2019

  109. [117]

    Why Compete When You Can Work Together: FPGA-ASIC Integration for Persistent RNNs,

    E. Nurvitadhi et al., "Why Compete When You Can Work Together: FPGA-ASIC Integration for Persistent RNNs," 2019 IEEE 27th Annual International Symposium on Field -Programmable Custom Computing Machines (FCCM), San Diego, CA, USA, 2019, pp. 199- 207, doi: 10.1109/FCCM.2019.00035

    2019

  110. [118]

    Intel Stratix 10 NX FPGA: AI -Optimized FPGA for High-Bandwidth, Low -Latency AI Acceleration (SS -1121-2.0),

    Intel Corp., “Intel Stratix 10 NX FPGA: AI -Optimized FPGA for High-Bandwidth, Low -Latency AI Acceleration (SS -1121-2.0),” 2020

    2020

  111. [119]

    Tensor Streaming Architecture Delivers Unmatched Performance for Compute Intensive Workloads,

    Dale Southard, “Tensor Streaming Architecture Delivers Unmatched Performance for Compute Intensive Workloads,” 2019

    2019

  112. [120]

    Dissecting the Graphcore IPU Architecture via Microbenchmarking,

    Z. Jia et al., “Dissecting the Graphcore IPU Architecture via Microbenchmarking,”arXiv preprint arXiv:1912.03413, 2019

    1912 arXiv

  113. [121]

    Think Fast: A Tensor Streaming Processor (TSP) for Accelerating Deep Learning Workloads,

    D. Abts et al., “Think Fast: A Tensor Streaming Processor (TSP) for Accelerating Deep Learning Workloads,” in International Symposium on Computer Architecture (ISCA), 2020, pp. 145–158 Omar Alhabshi is a professional engineer at ASEA Brown Boveri (ABB) Transmission & Distribut...

    2020

  114. [2008]

    General-Purpose Processing using Graphics Processing Units

  115. [2024]

    He has led MRI -based radiomics projects in hepatocellular carcinoma and published on the applications of large language models in medical tasks and cancer imaging

    His rese arch focuses on liver and abdominal MRI, with emphasis on radiomics, quantitative imaging biomarkers, and artificial intelligence in oncology. He has led MRI -based radiomics projects in hepatocellular carcinoma and published on the applications of large language mode...

    1985

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.