Pith.

Medical Physics · arXiv:2611.17211 v1 · submitted 12 Nov 2026 · 47 references

A Feasibility Study of Whole-Cerebrum Task-Based fMRI at 0.55 T: Motor and Visual Tasks Using Single-Shot EPI

Parsa Razmara, Takfarinas Medani, Anand A. Joshi, Majid Abbasi Sisara, Ye Tian, Sophia X. Cui, Justin P. Haldar, Krishna S. Nayak

show 1 more author

Richard M. Leahy

The paper says

“Purpose: To evaluate the feasibility of whole-cerebrum task-based fMRI on a modern 0.55T system, examining the effects of extended scan duration and within-session consistency on functional mapping outcomes and comparison to acquisition with matched tasks at 3.0T.”

more of the abstract

Methods: Six healthy participants were scanned at 0.55T using a single-shot EPI sequence during bilateral finger-tapping and visual checkerboard tasks. Data were processed using an open-source pipeline. To investigate sensitivity improvements, two consecutive 5-minute runs were concatenated. Within-session consistency was also assessed. A representative 3T dataset provided additional reference context for temporal dynamics and statistical sensitivity. Results: Task-evoked activations were detected at 0.55T using Family-Wise Error (FWE) rate correction (p < 0.05) in both motor and visual cortices. While single 5-minute runs provided baseline detectability, extending the acquisition to 10 minutes increased t-statistic magnitudes and cluster extents while substantially reducing background noise (47% reduction in voxelwise dispersion). Run-to-run comparisons showed consistent peak locations and statistical significance across participants. Comparison with the 3T reference demonstrated that statistical significance increases with acquisition time so that a 0.55T scan approximately matches 3T significance levels at twice the scan duration across the range compared. Conclusion: Whole-cerebrum task fMRI is feasible at 0.55T using standard EPI and modern high-performance gradients. These results indicate that modest extensions in scan duration can compensate for reduced intrinsic BOLD sensitivity at lower field strengths.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

Holds up, with conditions. Honest, useful feasibility evidence that 0.55T single-shot EPI can map motor and visual cortex with FWE-corrected significance; the 2x duration ratio is preliminary and presented too firmly. This line is wrong

4 major objections · 4 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 47 references say.

The paper says, at Results, Figure 1C and Figure S1

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.67):

“The duration-ratio estimates should therefore be interpreted as preliminary empirical values rather than calibrated benchmarks.”

Pith says: Objection 1 of 4 · major

The approximately-2× duration ratio is the paper's most load-bearing quantitative claim, and it is presented without any uncertainty quantification. The ratio is read off a 3T curve from one participant and a 0.55T group-mean curve from four participants (8 runs), yet no confidence interval, bootstrap, or cross-participant spread is reported for the ratio itself.

more of objection 1

Figure S2 shows substantial run-to-run variability in the p-value curves, so the reader cannot tell whether the true ratio is 1.5×, 2×, or 3×. The authors do state (b73, b119) that the estimates are preliminary, which is appropriate, but the abstract and conclusions restate the ratio as an approximate fact. I request that the authors add a bootstrap or per-participant summary of the duration ratio (e.g., range or CI across the five 0.55T participants, and across reasonable truncation points), or explicitly demote the ratio from the abstract to a qualitative statement such as 'on the order of 2×'.

The paper says, at Methods, cumulative GLM analysis

The review did not quote this passage.

Pith says: Objection 2 of 4 · major

The cumulative-sensitivity analysis averages the signal from a 10-vertex neighborhood centered on the peak motor vertex selected from the 0.55T full-run data. Because the vertex is chosen after seeing the full data, the t-statistic growth at short truncation lengths is optimistically biased relative to a pre-specified ROI, and this bias may differ between 0.55T and 3T if the 3T vertex selection used the 0.55T location.

more of objection 2

The authors should state explicitly how the 3T analysis region was chosen and, if possible, confirm that the qualitative 2× result persists when a fixed anatomical ROI (e.g., Brodmann area 4 or a hand-knob mask) is used instead of the peak-selected vertex.

The paper says, at Methods, duration-ratio computation

The review did not quote this passage.

Pith says: Objection 3 of 4 · major

The description of how the duration ratio is computed is incomplete. The text says p-value curves 'were interpolated in log-space to compute the average duration ratio' but does not define the reference p-value level at which the horizontal comparison is made, how the average is taken across participants or truncation points, or how the group p-value in Figure S1 is derived from the mean t-statistic across runs.

more of objection 3

Without this definition, the reported ratio of approximately 2 is not reproducible from the text alone. Please specify the exact computation, including the reference threshold and the averaging procedure, in the Methods section.

The paper says, at Results, supplementary tables

The review did not quote this passage.

Pith says: Objection 4 of 4 · major

The manuscript states that 'one of the six participants' was excluded from the finger-tapping analysis and that the multi-subject mean in Figure S1 covers '4 participants, 8 runs total,' with Figure S2 showing runs from five participants (Vol1-Vol5). The relationship between these numbers is not fully spelled out: which participant contributed to the 5-participant individual-run figure but not to the 4-participant mean, and why does Figure S1 exclude a second participant despite the main-text exclusion being only one?

more of objection 4

Please clarify the participant accounting in the main text or figure captions.

4 minor objections

The paper says, at Abstract

Pith says: minor objection 1 of 4

The abstract states the 2× duration ratio as a result without the qualifying language that appears in the main text (b73, b119). Consider adding 'approximately' and 'preliminary' to the abstract sentence to match the body's hedged framing.

The paper says, at Table 1

Pith says: minor objection 2 of 4

The table reports 3T in-plane resolution of 2.8 mm but the text says 'both used the 3T center's standard neuroscience protocols' while the 0.55T protocol uses no parallel imaging. Please state explicitly in the text or table caption that the 3T scan used GRAPPA R=2, since the difference in acceleration between field strengths is relevant to interpreting the sensitivity comparison.

show 2 more

The paper says, at Figure 1 caption

Pith says: minor objection 3 of 4

The caption states that t-statistics were computed from a '10-vertex neighborhood centered on the peak motor vertex displayed in (A)' but does not specify whether the same vertex was used for the 3T curve. Please clarify.

The paper says, at General

Pith says: minor objection 4 of 4

The sentence 'These values are preliminary empirical estimates from a small cohort rather than definitive scaling factors' (b73) is good practice; the same caveat should appear in the conclusion, which currently states the duration-compensation claim without caveat.

What would settle it

Recompute the duration ratio from the 0.55T data using a bootstrap: for each of the 4 participants (8 runs), draw with replacement 1000 resamples of the cumulative t-statistic curves, compute the ratio of times at which the 0.55T p-value crosses the 3T p-value at each truncation, and report the 95% CI of the median ratio. If the CI spans [1.5, 3.0] or wider, the ratio is too uncertain to support the headline claim without qualification. If the CI is tight around 2.0, the concern is mitigated.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 13 Nov 2026

  1. Reader holds up, with conditions confidence moderate
  2. Skeptic agrees 9 s · 15,265 tokens read, 745 written
  3. Editor wrote the verdict above 24 s · 16,204 tokens read, 3,168 written
  4. Referee major revision: 4 major objections, 4 minor 16 s · 15,717 tokens read, 1,671 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

Bottom line: this is a solid, honest feasibility study, and the 2x duration-ratio claim is the only place where the packaging exceeds the evidence. If you work on low-field fMRI, read it; if you are comparing it to 3T benchmarks, treat the ratio as suggestive.

What is actually new: it is one of the few systematic demonstrations of whole-cerebrum task fMRI on a modern 0.55T scanner using ordinary single-shot EPI, with motor and visual tasks, FWE-corrected statistics, run-to-run consistency, and a cumulative analysis showing sensitivity growth with scan time. The single 5-minute runs already produce FWE-significant activation in M1/SMA and V1; 10-minute concatenation raises peak t values and cuts background noise. That is a useful engineering benchmark, not a conceptual advance, but it is exactly what the low-field MRI community needs more of.

Where the paper is soft: the headline ratio (0.55T needs roughly 2x scan duration to match 3T) is derived from one 3T participant and a group mean of four 0.55T participants (eight runs), and there is no confidence interval, bootstrap, or sensitivity analysis around it. The authors explicitly call it preliminary (sentences b73, b119), and Figure S2 shows real run-to-run scatter, so the ratio could easily be 1.5x or 3x. The abstract and conclusion state it as an approximate fact (sentence b11), which is too strong for n=1 at 3T. I would fix that language and add some uncertainty quantification, even a simple bootstrap over runs. The rest of the limitations (small cohort, single-subject 3T reference, simple tasks, unoptimized TE/TR) are stated plainly in the paper, so I will not re-litigate them.

I do not see a circularity problem. The t-statistic as a function of duration is an empirical curve; the sqrt(N) envelope is fitted descriptively, not used to predict an independent outcome. There is no invented machinery.

Who should read it: people planning low-field or portable fMRI studies, and reviewers of similar feasibility papers. It is not a clinical validation or a cognitive neuroscience result.

My take: send it out. It deserves a serious referee. The substantive claims for simple motor and visual mapping at 0.55T hold up; the needed revisions are presentation, uncertainty quantification on the duration ratio, and a clearer line between feasibility evidence and calibrated scaling.

The strongest objection

The paper's strongest quantitative claim is the 2× duration ratio (abstract, Fig. 1C, Fig. S1). The derivation uses a single 3T participant (n=1) and a 0.55T group mean from 4 participants (8 runs). The authors explicitly call these 'preliminary empirical estimates' (sentence b73) and 'not definitive scaling factors' (b119). However, the abstract and conclusion state the ratio as an approximate fact ('approximately matches 3T significance levels at twice the scan duration'). The problem is not that the ratio is stated—it is that no confidence interval, bootstrap, or uncertainty quantification is provided. The ratio could be 1.5× or 3× depending on which participant or which truncation point is chosen. The paper's own Fig. S2 shows individual-run p-value curves that vary substantially; the ratio is read off the group-mean curve. Without a measure of uncertainty, a reader cannot assess whether the ratio is robust. This is the load-bearing concern because the entire practical message—that 0.55T can compensate for lower SNR by doubling scan time—hinges on this number being roughly correct. If the true ratio is larger, the feasibility argument weakens; if smaller, it strengthens. The paper does not provide the statistical machinery to decide.

The reader's weakest_assumption correctly identifies that the 3T comparison is single-subject and the duration ratio is derived from that single comparison plus a multi-subject 0.55T analysis. My concern is a more precise version of the same issue: the ratio is presented without uncertainty quantification, which is the specific technical gap. The paper's own limitations (b73, b119) acknowledge the preliminary nature, but the abstract and conclusion present the ratio as a finding. The concrete test I propose (bootstrap CI) would directly address whether the ratio is stable. I agree with the reader's verdict (CONDITIONAL) and see no reason to change it—the paper is a reasonable feasibility study, and the concern does not invalidate the core feasibility claim, only the precision of the quantitative ratio. Hence verdict_should_be UNCHANGED. The agreement_with_reader is 'agree' because we identify the same weak point.

The referee's report

If the results hold, the paper provides useful proof-of-concept evidence for a practically relevant claim: that modern 0.55T scanners with high-performance gradients can produce usable whole-cerebrum task fMRI with standard single-shot EPI, and that extended scan duration can partially compensate for reduced BOLD sensitivity. The study is methodologically clean in several respects: the feasibility claim rests on FWE-corrected first-level statistics, the main quantitative claims are empirically grounded rather than fitted to data, the cumulative-sensitivity analysis follows the expected sqrt(N) trend, and the authors are unusually explicit about the preliminary nature of the duration ratio. The single-subject 3T reference is acknowledged as contextual rather than a calibrated benchmark. The paper's main value is enabling researchers to plan 0.55T task fMRI studies; this value depends on the robustness of the approximately-2× duration ratio, which is the weakest-supported quantitative claim.

Pith's reading of the paper (a machine's summary, not the authors')

This paper tries to establish that modern low-field MRI can perform genuine whole-brain functional mapping, not just anatomy. On a 0.55T scanner with high-performance gradients, using an ordinary single-shot echo-planar imaging (EPI) sequence and an open-source processing pipeline, the authors detect statistically significant activations (family-wise-error corrected $p<0.05$) in motor and visual cortex of all included participants during finger-tapping and checkerboard tasks. They show that extending the acquisition from 5 to 10 minutes raises mean peak $t$-statistics by roughly a third, enlarges clusters two- to threefold, and cuts spurious background correlation sharply. Comparing cumulative significance curves against a 3T reference, they estimate that a 0.55T scan takes about twice as long as a 3T scan to reach the same significance level. If that holds, low-field scanners become a credible platform for task fMRI in settings where high field is impractical, such as patients with implants, combined EEG-fMRI recordings, and lower-cost imaging.

What carries the argument. The argument is carried by three working parts: single-shot gradient-echo EPI with echo time matched to the low-field gray-matter $T_2^*$ (TE = 85 ms), a block-design paradigm modeled with the canonical hemodynamic response function in a voxelwise GLM with family-wise-error correction, and a cumulative truncation analysis that recomputes the $t$-statistic on a 10-vertex neighborhood of the peak motor vertex as scan duration grows, extracting the duration ratio by interpolating 0.55T and 3T $p$-value curves in log-space. Concatenating two 5-minute runs is the experimental lever that demonstrates sensitivity accumulating with time, and the fitted $\sqrt{N}$ envelopes tie that accumulation to the shrinking standard error of the GLM estimate as time points are added.

Core claim. The central claim is that the blood-oxygenation-level-dependent (BOLD) contrast mechanism at 0.55T is strong enough for whole-cerebrum, block-design task fMRI when the acquisition is matched to low-field physics. With the echo time set near the gray-matter $T_2^*$ at 0.55T (TE = 85 ms), no parallel imaging or multiband, and a voxelwise GLM with family-wise-error (FWE) correction, a single 5-minute run produced significant bilateral activation in primary motor cortex (M1) and supplementary motor area for finger-tapping, and in primary visual cortex (V1) for a flickering checkerboard, in every included participant. Concatenating two runs into 10 minutes increased mean peak $t$-scores from 9.56 to 12.68 for motor and 7.18 to 9.24 for visual, grew clusters roughly 2- to 3.5-fold, and reduced voxelwise dispersion of non-target correlations by 47%, shifting the maps toward smoother, more spatially specific activation. A cumulative re-analysis of the $t$-statistic as a function of scan duration, fitted to the expected $\sqrt{N}$ growth, shows 0.55T reaching 3T-equivalent $p$-values at roughly twice the scan duration; the authors present this ratio as a preliminary empirical estimate, not a calibrated benchmark.

The premise it rests on. The central claim that 0.55T needs about twice the scan duration to match 3T significance rests on a single 3T participant compared with a small 0.55T cohort, so the 2x ratio could shift with more subjects, other tasks, or protocol differences.

If the paper is right (Pith's inferences, not the authors')

Whole-cerebrum task fMRI at 0.55T does not require specialized readouts or acceleration: a standard single-shot EPI sequence and an open-source pipeline detect corrected-significant activation in primary sensorimotor and visual cortex.

Protocol planning at 0.55T can budget approximately twice the scan duration of a 3T study for strong block-design tasks and expect comparable FWE-corrected significance.

Within-session run-to-run consistency of peak locations and significance supports repeated or longitudinal functional mapping within a single 0.55T session.

Longer acquisitions at 0.55T improve map quality beyond peak statistics, suppressing background spurious correlation and yielding smoother, more spatially specific activation maps.

The reduced geometric distortion at 0.55T makes frontal and temporal regions a natural next target for paradigms whose BOLD responses are weaker than primary-cortex responses.

The 2x duration ratio is calibrated on strong primary-cortex tasks; it is a testable question whether weaker or higher-order paradigms need a larger ratio, which would raise the practical cost of low-field fMRI for cognitive studies.

The paper motivates low-field fMRI through EEG compatibility and implant patients but measures neither; a direct simultaneous EEG-fMRI recording at 0.55T would test that promised payoff.

Because the sensitivity gain comes partly from reduced variance (lower non-target dispersion) rather than only from more signal, systematic TE/TR optimization might push the required duration ratio below 2.

A matched multi-subject 3T arm is the natural next experiment, since the current ratio rests on one 3T participant and could partly reflect participant or protocol differences rather than field strength alone.

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

Purpose: To evaluate the feasibility of whole-cerebrum task-based fMRI on a modern 0.55T system, examining the effects of extended scan duration and within-session consistency on functional mapping outcomes and comparison to acquisition with matched tasks at 3.0T. Methods: Six healthy participants were scanned at 0.55T using a single-shot EPI sequence during bilateral finger-tapping and visual checkerboard tasks. Data were processed using an open-source pipeline. To investigate sensitivity improvements, two consecutive 5-minute runs were concatenated. Within-session consistency was also assessed. A representative 3T dataset provided additional reference context for temporal dynamics and statistical sensitivity. Results: Task-evoked activations were detected at 0.55T using Family-Wise Error (FWE) rate correction (p < 0.05) in both motor and visual cortices. While single 5-minute runs provided baseline detectability, extending the acquisition to 10 minutes increased t-statistic magnitudes and cluster extents while substantially reducing background noise (47% reduction in voxelwise dispersion). Run-to-run comparisons showed consistent peak locations and statistical significance across participants. Comparison with the 3T reference demonstrated that statistical significance increases with acquisition time so that a 0.55T scan approximately matches 3T significance levels at twice the scan duration across the range compared. Conclusion: Whole-cerebrum task fMRI is feasible at 0.55T using standard EPI and modern high-performance gradients. These results indicate that modest extensions in scan duration can compensate for reduced intrinsic BOLD sensitivity at lower field strengths.

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-11-14

The authors' main claim, as the classifier picked it out 0.99

“Conclusion: Whole-cerebrum task fMRI is feasible at 0.55T using standard EPI and modern high-performance gradients.”

How the main claim is stated: asserted 1.00

What the abstract itself states, as probabilities: claims first 0.03 · states limitation 0.33 · human participants 0.99 · controlled comparison 0.60 · randomized assignment 0.02 · states quantitative result 0.99 · compares with prior results 0.96 · mentions code or data availability 0.15

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
Purpose: To evaluate the feasibility of whole-cerebrum task-based fMRI on a modern 0.55T system, examining the effects of extended scan duration and within-session consistency on functional mapping outcomes and comparison to acquisition with matched tasks at 3.0T. contribution 0.85 0.44 0.03 unclear 0.66
Methods: Six healthy participants were scanned at 0.55T using a single-shot EPI sequence during bilateral finger-tapping and visual checkerboard tasks. method 1.00 0.96 0.01 asserted 0.96
Data were processed using an open-source pipeline. method 1.00 0.80 0.01 asserted 0.99
To investigate sensitivity improvements, two consecutive 5-minute runs were concatenated. method 0.99 0.94 0.02 asserted 0.78
Within-session consistency was also assessed. method 0.99 0.89 0.01 asserted 0.81
A representative 3T dataset provided additional reference context for temporal dynamics and statistical sensitivity. method 0.98 0.51 0.01 asserted 0.94
Results: Task-evoked activations were detected at 0.55T using Family-Wise Error (FWE) rate correction (p < 0.05) in both motor and visual cortices. result 1.00 0.96 0.04 asserted 1.00
While single 5-minute runs provided baseline detectability, extending the acquisition to 10 minutes increased t-statistic magnitudes and cluster extents while substantially reducing background noise (47% reduction in voxelwise dispersion). result 1.00 0.96 0.06 asserted 1.00
Run-to-run comparisons showed consistent peak locations and statistical significance across participants. result 1.00 0.95 0.02 asserted 1.00
Comparison with the 3T reference demonstrated that statistical significance increases with acquisition time so that a 0.55T scan approximately matches 3T significance levels at twice the scan duration across the range compared. result 0.97 0.90 0.13 asserted 0.99
Conclusion: Whole-cerebrum task fMRI is feasible at 0.55T using standard EPI and modern high-performance gradients. interpretation 0.75 0.82 0.95 asserted 1.00
These results indicate that modest extensions in scan duration can compensate for reduced intrinsic BOLD sensitivity at lower field strengths. interpretation 1.00 0.83 0.18 hedged possibility 0.99

What kind of paper the classifier reports this is: empirical study 1.00 · evidence from observation or experiment 1.00

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-14: 1 request, 8,875 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

Figures (4)
Figure 1 of the paper
Figure 1. BOLD signal time series and statistical sensitivity. (A-B) Representative single-vertex task fMRI time series at 0.55T and 3T from a single participant. The plotted vertex was selected based on the maximum correlation with the HRF-convolved reference waveform in the 0.55T data within (A) the motor cortex ROI and (B) the V1 ROI. The corresponding anatomical location was extracted for 3T comparison. To facilitate visual comparison of temporal dynamics between field strengths, time series were z-score normalized and overlaid with the reference waveform (dashed line). Shaded regions mar
Figure 2 of the paper
Figure 2. Statistical activation maps for the finger-tapping and visual tasks at 0.55T. Representative whole-cerebrum t-statistic maps from a single participant, derived from concatenated 10-minute runs and overlaid on the 0.55T T1-weighted anatomical image. (A) Finger-tapping task, showing significant activation in the primary motor cortex; subject-specific statistics (peak t-scores, p-values, MNI coordinates, and cluster sizes) are in Table S2, and selected coronal slices in Figure S3. (B) Visual task, showing significant contiguous activation in the primary visual cortex (V1) and adjacent
Figure 3 of the paper
Figure 3. 0.55T and 3T fMRI BOLD activation results for a single participant (Motor Task), overlaid on the T1-weighted anatomical images acquired at the respective field strengths. (A) t-statistic maps for the finger-tapping experiment acquired at 0.55T. Peak statistics include t=19.2 (MNI: [−44,−10,58], cluster size: 2,070 voxels) and t=16.8 (MNI: [34,−26,62], cluster size: 2,107 voxels). (B) Corresponding t-statistic maps acquired at 3T, provided as a reference to evaluate spatial localization. Peak statistics at 3T include t=28.4 (MNI: [42,−14,53], cluster size: 2,013 voxels) and t=24.9 (M
Figure 4 of the paper
Figure 4. 0.55T and 3T fMRI BOLD activation results for a single participant (Visual Task), overlaid on the T1-weighted anatomical images acquired at the respective field strengths. (A) t-statistic maps for the visual experiment acquired at 0.55T. Peak statistics include t=11.0 (MNI: [6, -89, -8], cluster size: 302 voxels). (B) Corresponding t-statistic maps acquired at 3T. Peak statistics at 3T include t=13.4 (MNI: [-11, -92, -7], cluster size: 2,781 voxels). All reported peaks are significant at p < 0.001 (FWE-corrected). Maps are thresholded at p < 0.05 (FWE-corrected) and displayed at con

Figures are the authors', reproduced from the paper.

References (47)
  1. [1] Bandettini PA. Twenty years of functional MRI: the science and the stories. Neuroimage. 2012;62(2):575 –588 on Pith
  2. [2] Thulborn KR. Clinical rationale for very-high-field (3.0 Tesla) functional magnetic resonance imaging. Top Magn Reson Imaging. 1999;10(1):37–50 on Pith
  3. [3] Duyn JH. The future of ultra-high field MRI and fMRI for study of the human brain. Neuroimage. 2012;62(2):1241–1248 on Pith
  4. [4] De Martino F, Yacoub E, Kemper VG, et al. The impact of ultra-high field MRI on cognitive and computational neuroimaging. Neuroimage. 2018;168:366–382. doi:10.1016/j.neuroimage.2017.03.060 on Pithdoi:10.1016/j.neuroimage.2017.03.060
  5. [5] Ugurbil K. Imaging at ultrahigh magnetic fields: history, challenges, and solutions. Neuroimage. 2018;168:7 –32. doi:10.1016/j.neuroimage.2017.07.007 on Pithdoi:10.1016/j.neuroimage.2017.07.007
  6. [6] Campbell-Washburn AE, Ramasawmy R, Restivo MC, et al. Opportunities in interventional and diagnostic imaging by using high-performance low-field-strength MRI. Radiology. 2019;293(2):384–393 on Pith
  7. [7] Srivastava G, Crottaz-Herbette S, Lau KM, Glover GH, Menon V. ICA-based procedures for removing ballistocardiogram artifacts from EEG data acquired in the MRI scanner. Neuroimage. 2005;24(1):50 –60 on Pith
  8. [8] Razmara P, Jeong W, Kommineni A, Cassani R, Leahy RM, Medani T. Best practices in EEG analysis: preprocessing, modeling, and machine learning. arXiv preprint. 2026. doi:10.48550/arXiv.2609.36609 on PithPith's reviewarXivdoi:10.48550/arxiv.2609.36609
  9. [9] Lee J, Razmara P, Huang K, et al. An approach to simultaneous acquisition of real -time MRI video, EEG, and surface EMG for articulatory, brain, and muscle activity during speech production. arXiv preprint. 2026 on Pith
  10. [10] Sutton BP, Lam F. Imaging in the presence of magnetic field inhomogeneities. In: Akcakaya M, Doneva MI, Prieto C, eds. Magnetic Resonance Image Reconstruction: Theory, Methods and Applications. Advances in Magnetic Resonance Technology and Applications. Vol 7. Academic Press; ... on Pithdoi:10.1016/b978-
  11. [11] Warbrick T. Simultaneous EEG-fMRI: what have we learned and what does the future hold? Sensors (Basel). 2022;22(6):2262. doi:10.3390/s22062262 on Pithdoi:10.3390/s22062262
  12. [12] Razmara P, Medani T, Abbasi Sisara M, et al. Feasibility of simultaneous EEG-fMRI at 0.55 T: recording, denoising, and functional mapping. arXiv preprint. 2026. doi:10.48550/arXiv.2602.13489 on PithPith's reviewarXivdoi:10.48550/arxiv.2602.13489
  13. [13] Sarracanie M, LaPierre CD, Salameh N, Waddington DEJ, Witzel T, Rosen MS. Low -cost high-performance MRI. Sci Rep. 2015;5:15177. doi:10.1038/srep15177 on Pithdoi:10.1038/srep15177
  14. [14] Marques JP, Simonis FFJ, Webb AG. Low-field MRI: an MR physics perspective. J Magn Reson Imaging. 2019;49(6):1528–1542. doi:10.1002/jmri.26637 on Pithdoi:10.1002/jmri.26637
  15. [15] Wald LL, McDaniel PC, Witzel T, Stockmann JP, Cooley CZ. Low-cost and portable MRI. J Magn Reson Imaging. 2020;52(3):686–696. doi:10.1002/jmri.26942 on Pithdoi:10.1002/jmri.26942
  16. [16] Marques JP, van Kemenade W, Gazzo S, Grodzki D, Knopp EA, Stainsby J. ESMRMB annual meeting roundtable discussion: "when less is more: the view of MRI vendors on low -field MRI". MAGMA. 2021;34(4):479–482. doi:10.1007/s10334-021-00938-9 on Pithdoi:10.1007/s10334-021-00938-9
  17. [17] Arnold TC, Freeman CW, Litt B, Stein JM. Low-field MRI: clinical promise and challenges. J Magn Reson Imaging. 2023;57(1):25–44. doi:10.1002/jmri.28408 on Pithdoi:10.1002/jmri.28408
  18. [18] Campbell-Washburn AE, Keenan KE, Hu P, et al. Low-field MRI: a report on the 2022 ISMRM workshop. Magn Reson Med. 2023;90(4):1682–1694. doi:10.1002/mrm.29743 on Pithdoi:10.1002/mrm.29743
  19. [19] Wang Y, van Gelderen P, de Zwart JA, Campbell-Washburn AE, Duyn JH. FMRI based on transition-band balanced SSFP in comparison with EPI on a high-performance 0.55 T scanner. Magn Reson Med. 2021;85(6):3196–3210. doi:10.1002/mrm.28657 on Pithdoi:10.1002/mrm.28657
  20. [20] Halder A, Harris CT, Wiens CN, Soddu A, Chronik BA. Optimization of gradient-echo echo-planar imaging for T2* contrast in the brain at 0.5 T. Sensors (Basel). 2023;23(20):8428. doi:10.3390/s23208428 on Pithdoi:10.3390/s23208428
show 27 more
  1. [21] Razmara P, Medani T, Joshi AA, et al. A feasibility study of task-based fMRI at 0.55 T. Proc Intl Soc Magn Reson Med. 2025;33:0491. doi:10.58530/2025/0491 on Pithdoi:10.58530/2025/0491
  2. [22] Joshi AA, Cui SX, Chan CC, et al. An evaluation of brain volume and cortical thickness measurement at 0.55 T. MAGMA. 2026. doi:10.1007/s10334-026-01380-5 on Pithdoi:10.1007/s10334-026-01380-5
  3. [23] Kung HT, Cui SX, Kaplan JT, et al. Diffusion tensor brain imaging at 0.55T: a feasibility study. Magn Reson Med. 2024;92(4):1649–1657. doi:10.1002/mrm.30156 on Pithdoi:10.1002/mrm.30156
  4. [24] Gati JS, Menon RS, Uǧurbil K, Rutt BK. Experimental determination of the BOLD field strength dependence in vessels and tissue. Magn Reson Med. 1997;38(2):296–302. doi:10.1002/mrm.1910380220 on Pithdoi:10.1002/mrm.1910380220
  5. [25] Fera F, Yongbi MN, van Gelderen P, Frank JA, Mattay VS, Duyn JH. EPI-BOLD fMRI of human motor cortex at 1.5 T and 3.0 T: sensitivity dependence on echo time and acquisition bandwidth. J Magn Reson Imaging. 2004;19(1):19–26. doi:10.1002/jmri.10440 on Pithdoi:10.1002/jmri.10440
  6. [26] Telesford QK, Gonzalez-Moreira E, Xu T, et al. An open-access dataset of naturalistic viewing using simultaneous EEG-fMRI. Sci Data. 2023;10:554. doi:10.1038/s41597-023-02458-8 on Pithdoi:10.1038/s41597-023-02458-8
  7. [27] Shattuck DW, Leahy RM. BrainSuite: an automated cortical surface identification tool. Med Image Anal. 2002;6(2):129–142. doi:10.1016/S1361-8415(02)00054-3 on Pithdoi:10.1016/s1361-8415(02)00054-3
  8. [28] Joshi AA, McCoy D, Chong M, et al. BFP: BrainSuite fMRI pipeline. In: Proc 24th Annual Meeting of the Organization for Human Brain Mapping (OHBM); 2018; Singapore on Pith
  9. [29] Friston KJ, Holmes AP, Worsley KJ, Poline JP, Frith CD, Frackowiak RSJ. Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Mapp. 1994;2(4):189 –210 on Pith
  10. [30] Ashburner J, Friston KJ. Unified segmentation. Neuroimage. 2005;26(3):839 –851. doi:10.1016/j.neuroimage.2005.02.018 on Pithdoi:10.1016/j.neuroimage.2005.02.018
  11. [31] Bandettini PA, Jesmanowicz A, Wong EC, Hyde JS. Processing strategies for time -course data sets in functional MRI of the human brain. Magn Reson Med. 1993;30(2):161–173. doi:10.1002/mrm.1910300204 on Pithdoi:10.1002/mrm.1910300204
  12. [32] Birn RM, Molloy EK, Patriat R, et al. The effect of scan length on the reliability of resting -state fMRI connectivity estimates. Neuroimage. 2013;83:550–558. doi:10.1016/j.neuroimage.2013.05.099 on Pithdoi:10.1016/j.neuroimage.2013.05.099
  13. [33] Poldrack RA, Fletcher PC, Henson RN, Worsley KJ, Brett M, Nichols TE. Guidelines for reporting an fMRI study. Neuroimage. 2008;40(2):409–414. doi:10.1016/j.neuroimage.2007.11.048 on Pithdoi:10.1016/j.neuroimage.2007.11.048
  14. [34] Karjalainen T, Seppälä K, Glerean E, et al. Opioidergic regulation of emotional arousal: a combined PET –fMRI study. Cereb Cortex. 2019;29(9):4006–4016. doi:10.1093/cercor/bhy281 on Pithdoi:10.1093/cercor/bhy281
  15. [35] Abraham A, Pedregosa F, Eickenberg M, et al. Machine learning for neuroimaging with scikit -learn. Front Neuroinform. 2014;8:14. doi:10.3389/fninf.2014.00014 on Pithdoi:10.3389/fninf.2014.00014
  16. [36] Triantafyllou C, Wald LL, Hoge RD. Echo-time and field strength dependence of BOLD reactivity in veins and parenchyma using flow-normalized hypercapnic manipulation. PLoS One. 2011;6(9):e24519 on Pith
  17. [37] Miller KL, Smith SM, Jezzard P, Pauly JM. High-resolution FMRI at 1.5T using balanced SSFP. Magn Reson Med. 2006;55(1):161–170. doi:10.1002/mrm.20753 on Pithdoi:10.1002/mrm.20753
  18. [38] van der Zwaag W, Francis S, Head K, et al. fMRI at 1.5, 3 and 7 T: characterising BOLD signal changes. Neuroimage. 2009;47(4):1425–1434. doi:10.1016/j.neuroimage.2009.05.015 on Pithdoi:10.1016/j.neuroimage.2009.05.015
  19. [39] Gorno-Tempini ML, Hutton C, Josephs O, Deichmann R, Price C, Turner R. Echo time dependence of BOLD contrast and susceptibility artifacts. Neuroimage. 2002;15(1):136–142. doi:10.1006/nimg.2001.0967 on Pithdoi:10.1006/nimg.2001.0967
  20. [40] Haacke EM, Brown RW, Thompson MR, Venkatesan R. Magnetic Resonance Imaging: Physical Principles and Sequence Design. Wiley-Liss; 1999 on Pith
  21. [41] Fleischer V, Muthuraman M, Anwar AR, et al. Continuous reorganization of cortical information flow in multiple sclerosis: a longitudinal fMRI effective connectivity study. Sci Rep. 2020;10:806. doi:10.1038/s41598-020-57895-x on Pithdoi:10.1038/s41598-020-57895-x
  22. [42] Barth M, Breuer F, Koopmans PJ, Norris DG, Poser BA. Simultaneous multislice (SMS) imaging techniques. Magn Reson Med. 2016;75(1):63–81. doi:10.1002/mrm.25897 on Pithdoi:10.1002/mrm.25897
  23. [43] Narsude M, Gallichan D, van der Zwaag W, Gruetter R, Marques JP. Three-dimensional echo planar imaging with controlled aliasing: a sequence for high temporal resolution functional MRI. Magn Reson Med. 2016;75(6):2350–2361. doi:10.1002/mrm.25817 on Pithdoi:10.1002/mrm.25817
  24. [44] Torabi Y, Razmara P, Ajorlou H, Baraeinejad B. NeuroMambaLLM: dynamic graph learning of fMRI functional connectivity in autistic brains using Mamba and language model reasoning. arXiv preprint. 2026 on Pith
  25. [45] Iglesias JE, Schleicher R, Laguna S, et al. Quantitative brain morphometry of portable low -field-strength MRI using super-resolution machine learning. Radiology. 2023;306(3):e220522. doi:10.1148/radiol.220522 on Pithdoi:10.1148/radiol.220522
  26. [46] Manso Jimeno M, Ravi KS, Jin Z, Oyekunle D, Ogbole G, Geethanath S. ArtifactID: identifying artifacts in low - field MRI of the brain using deep learning. Magn Reson Imaging. 2022;89:42 –48 on Pith
  27. [47] Mašková B, Rožánek M, Gajdoš O, Karnoub E, Kamenský V, Donin G. Assessment of the diagnostic efficacy of low-field magnetic resonance imaging: a systematic review. Diagnostics. 2024;14(14):1564. doi:10.3390/diagnostics14141564. Figures and Tables Table 1. Acquisition parameter... on Pithdoi:10.3390/diagnostics14141564

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment