{"id":"aa8b0f44-9cc2-452d-9cfa-169c5067232c","arxiv_id":"2608.11117","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A systematic multi-field measurement of relaxation times, relaxivities, and permittivities for nine hydrogel/electrolyte materials and three oils used in MRI phantoms, with practical candidate recommendations.","lead":"Researchers measured T1 and T2 relaxation times and electrical properties of common MRI phantom materials (hydrogels, salt solutions, oils) at three clinical field strengths, then rated which materials make reliable test phantoms. The dataset helps labs build cheap, consistent MRI quality-assurance phantoms, especially for low-field MRI-guided radiation therapy.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The reported low-field complex permittivities rest on an unvalidated extrapolation below the dielectric probe's rated 200 MHz floor; a cross-check against a low-frequency-rated setup is needed before the phantom recipe tables are relied on.","rationale":"I read the paper as a measurement-and-characterization study whose utility depends on the reliability of the tabulated T1, T2, and permittivity values. The T1/T2 acquisition and fitting are conventional, cover the three stated fields directly, and show reasonable agreement with prior literature where comparisons are offered, so I do not see a load-bearing objection to the relaxation-time half of the central claim. The weakest point is exactly the one the reader identified: the dielectric probe data below 200 MHz are extrapolated rather than measured, and this is not a peripheral detail because 0.35 T corresponds to 14.71 MHz and 1.5 T to 63.89 MHz. The paper even gives itself a contradictory rating: the E4991A analyzer is said to be rated from 10 MHz, while the E5061A/probe combination is said to be rated only above 200 MHz, and there is no reported cross-validation between the two configurations. My concrete test would settle this by examining whether the Cole-Cole extrapolation from >= 200 MHz data agrees with direct low-frequency measurements. If it does not, the permittivity tables at low field—and the associated conclusions about high-dielectric hydrogels and low-dielectric oils—would need to be downgraded to extrapolations or re-measured. I do not see this as requiring rejection of the whole paper: the T1/T2 data and the relaxivity trends could still stand, and many permittivity conclusions at 1.5 and 3 T may survive if the affected entries are restricted or flagged. The reader's conditional verdict is therefore appropriate; my stress-test does not move it, but it does sharpen the condition under which the paper should be accepted.","tokens_in":21329,"tokens_out":4132,"duration_ms":39526,"concrete_test":"For a representative subset (deionized water, 0.9% saline, 5% xanthan gum, 1 mM CuSO4, 1 mM Mn(NO3)2, and canola oil), refit the Cole-Cole model using only permittivity data at frequencies >= 200 MHz, then compare the resulting epsilon_s and sigma_s with the full-range fits reported in Tables 3-15. Separately, remeasure the same samples with the E4991A configuration rated from 10 MHz, or an equivalent low-frequency probe, directly acquiring data down to 14 MHz. If the epsilon_s or sigma_s values at 14.71 MHz shift by more than the stated fit uncertainties, the reported low-field complex permittivities are not supported and the permittivity-related recommendations should be revised or labeled as extrapolated only.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central deliverable is a combined dataset of T1, T2, and complex permittivity for phantom materials at 0.35, 1.5, and 3 T. The T1/T2 measurements are directly acquired at the three fields and are internally consistent, but the complex permittivity claim is less secure. In the Methods Permittivities paragraph, the authors state that measurements ranged from 14 MHz to 1.5 GHz for hydrogels and MnNO3 using an Agilent 85070E probe with an E5061A network analyzer, and then state: 'We extrapolated the static dielectrics and electric conductivities down to 14 MHz since the dielectric probe is rated for a range from 200 MHz to 20 GHz with the network analyzer.' That means the data below 200 MHz, including the 0.35 T Larmor frequency (14.71 MHz) and part of the 1.5 T band, are extrapolations rather than measurements. The reported epsilon_s and sigma_s values in Tables 3-15, and the conclusions that hydrogels have high static dielectrics and oils have epsilon_s < 10, therefore depend on an unvalidated Cole-Cole extrapolation across a band where the probe is not specified. The paper also describes a second configuration, the E4991A impedance analyzer rated from 10 MHz to 3 GHz, but no cross-check is reported between the two configurations on the same samples. The internal contradiction between 'rated 10 MHz to 3 GHz' and 'rated 200 MHz to 20 GHz' is not resolved, and it is exactly the low-frequency regime that matters for low-field MRI phantoms.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports T1, T2, and complex permittivity measurements for a panel of MRI phantom constituents—seven hydrogels, three paramagnetic salt solutions, and three oils—at 0.35, 1.5, and 3 T, with concentrations varied per material. T1 and T2 were acquired with standard IR-EPI and CPMG sequences at the three field strengths and fitted to monoexponential models; relaxivities are obtained from linear concentration fits. Complex permittivities were measured with coaxial dielectric probes over roughly 14 MHz to 1–1.5 GHz and fitted to a Cole-Cole model to extract static dielectric constants and static conductivities. The authors use a power law in B0 to interpolate T1 to 0.55 T and extrapolate to 7 T, and they draw practical conclusions about which materials are suitable as MRI phantom components, including that sodium alginate, Miller's LB agar, and xanthan gum are good hydrogel candidates, Mn(NO3)2 has the highest relaxivities, and the three oils are suitable low-dielectric materials.","tokens_in":21759,"tokens_out":2116,"duration_ms":19654,"significance":"If the dataset is reliable, it fills a practical gap: phantom builders currently need to assemble relaxation times, relaxivities, and permittivities from scattered sources or from studies at a single field strength. The paper's direct, multi-field comparison of common, inexpensive, commercially available phantom materials, with reported fit uncertainties, is potentially useful for QA phantom design in MR-guided radiotherapy (0.35 T), clinical imaging (1.5 and 3 T), and low-field/high-field applications. The study is also a useful reminder that simple field-strength scaling assumptions for T1 and relaxivity do not hold for every material; several reported deviations, such as r2 decreasing with field for CuSO4 and r1 rising for NiCl2, are falsifiable observations that can guide future modeling. The T1/T2 measurements are the strongest part because they are direct measurements at the target field strengths, use standard sequences, and include comparisons to earlier work for CuSO4, NiCl2, and sodium alginate.","major_comments":[{"comment":"The two instrument specification statements are internally inconsistent, and this matters because the low-frequency credibility of the results is at stake. The text first says the Agilent E4991A impedance analyzer has a rated range of 10 MHz to 3 GHz, then later says the 85070E/E5061A network analyzer combination is rated for 200 MHz to 20 GHz. No cross-check between the two configurations on the same samples is reported, so the reader cannot tell which configuration, if any, supports the 14 MHz lower bound claimed for the measurements. Please clarify which analyzer was used for which samples, report the manufacturer's rated frequency range for each configuration, and provide a direct comparison measurement between the two setups over a common frequency band.","section":"Methods, Permittivities paragraph"},{"comment":"The 7 T extrapolation is shown to be nonphysical for Miller's LB agar within the paper itself. The Figure 1 caption notes that the extrapolated R1 at 7 T for 5% w/w is lower than R1 for 4% w/w, so the power-law extrapolation of Eq. (3) reverses the monotonic concentration ordering at 7 T. Since T1 interpolation/extrapolation is presented as a deliverable in the Highlights and is used to compute r1 values, the authors should either restrict the claimed validity of Eq. (3) to the measured range or provide a criterion for when the power-law extrapolation is reliable; as written, the paper acknowledges a counterexample to its own extrapolation procedure for one of the recommended hydrogels.","section":"Figure 1 caption and Table 3"},{"comment":"The 0.55 T and 7 T r1 values are not independent predictions. Equation (3), with two free parameters A and C, is fitted to exactly three measured field strengths (0.35, 1.5, 3 T), leaving zero degrees of freedom; the 0.55 T interpolation and the 7 T extrapolation are deterministic transformations of the three measured points, and the quoted r1 uncertainties do not propagate the fit uncertainty or the model-form uncertainty. This should be stated explicitly wherever interpolated/extrapolated r1 values appear, and the error bars on those values should be presented as lower bounds or omitted.","section":"Eq. (3) and relaxivity r1 values"}],"minor_comments":[{"comment":"Polyethylene glycol 8000 is described with molecular weight 62.07 g/mol, which is the molecular weight of the ethylene glycol monomer, not PEG 8000; the correct nominal molecular weight should be on the order of 8000 g/mol. Please correct the value or the label.","section":"Methods, Hydrogels paragraph"},{"comment":"The abstract contains an unbalanced parenthesis: 'low dielectric (εs<10) phantoms for high field applications) but with low conductivities.' Please remove the extra closing parenthesis.","section":"Abstract and Introduction"},{"comment":"For deionized water, the reported T2 values are 2051.72 ms at 0.35 T, 1755.22 ms at 1.5 T, and 2422.19 ms at 3 T; a non-monotonic field dependence of T2 for pure water is unexpected and deserves a brief comment on possible measurement or fitting effects, since T2 of free water is generally expected to be nearly field-independent.","section":"Table 14"},{"comment":"Several table captions state 'measured at room temperature 20°C' while the Methods report 21°C for relaxation and 21.5°C for DC conductivity; please harmonize the temperature notation so readers can reproduce the measurements.","section":"Tables 3–13"},{"comment":"The Data Availability statement says data may be provided upon request. Given the paper's goal of serving as a reference dataset, depositing the raw T1/T2 fits and permittivity spectra (e.g., in a repository) would substantially increase the utility and verifiability of the tables.","section":"Data Availability"}],"recommendation":"major_revision","confidential_remarks":"The paper's core T1/T2 dataset is likely sound and useful, but the complex permittivity component, which supports several of the headline conclusions (low-dielectric oils, high-dielectric hydrogels, conductivity trends), depends on data collected below the probe's rated frequency floor and on an unexplained discrepancy between two instrument specifications. The authors appear capable of fixing this by adding a cross-check measurement or by weakening the claims; I do not see a career-level or integrity concern, but the error is load-bearing for a central part of the manuscript. There is also a self-acknowledged counterexample to the 7 T extrapolation for Miller's LB agar, which should be handled in the revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Thanks for the Pith report. My take: this is a useful reference paper, not a breakthrough. The combined matrix of nine water-based materials plus three oils measured under one protocol at 0.35, 1.5, and 3 T is genuinely new, and the 0.35 T relaxation data fill a real gap for MRgRT phantom builders. The T1/T2 measurements look solid: standard IR-EPI and CPMG fits, reasonable uncertainties, and good agreement with prior studies where comparisons exist (e.g., CuSO4/NiCl2 at 0.5 T, CuSO4 at 1.5 T, paramagnetic solutions at 3 T). I also give them credit for reporting supplier, lot, and price details, which makes the recipes reproducible. The relaxivity patterns they report—NiCl2 r1 rising with field, CuSO4 r2 dropping—are worth noting even if the paper doesn't push the theory hard.\n\nThe soft spots are real but not fatal. The biggest is the complex permittivity data. The paper describes two set-ups: one with an E4991A rated to 10 MHz, one with an E5061A where they state the probe is rated 200 MHz to 20 GHz and they 'extrapolated' static dielectric and conductivity down to 14 MHz. That unresolved discrepancy matters because the 0.35 T Larmor frequency is 14.71 MHz, exactly in the extrapolated band. The DC conductivity measurements serve as a cross-check for σs and generally agree well, which helps. But there is no independent validation for εs at low frequency. A quick cross-check on one or two samples with a low-frequency bridge or four-electrode cell would settle it. As written, I would trust the T1/T2 tables but treat the reported εs values, especially for oils, as provisional.\n\nOther issues: no raw data or analysis code released, and the 7 T extrapolation is acknowledged to be nonphysical for one Miller's LB agar sample. The Mn(NO3)2 relaxivities at 3 T are degraded by signal dephasing and the paper flags the missing points. These are minor because the core dataset stands on direct measurements. Also, the abstract says 'T1 decreased with increasing concentration except for PEG'—true, but the effect for PEG is essentially zero; the reader should not expect a dramatic exception.\n\nWho is this for? Anyone building or validating MRI phantoms at low or clinical field strengths. It deserves a serious referee, but the referee should ask for clarification of the permittivity extrapolation and ideally a cross-check. I would not desk reject.","headline":"Useful multi-field phantom-material reference, solid on T1/T2, but the low-frequency permittivity extrapolation needs validation before the recipe tables are used uncritically.","tokens_in":22294,"tokens_out":3991,"would_cite":true,"duration_ms":32693,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Measurements of T1, T2, and complex permittivity for common MRI phantom materials at 0.35, 1.5, and 3 T support specific material recommendations for phantom builders.","keywords":["Relaxation time","dielectric strength","electrical conductivity","MRI","phantom liquids","hydrogels","paramagnetic salts","oils"],"falsifier":"Measure the same samples with a dielectric probe rated below 14 MHz, or with time-domain reflectometry, and compare the independently obtained $\\varepsilon_s$ and $\\sigma_s$ with the extrapolated values in Tables 3-15; disagreement beyond the reported uncertainties would overturn the permittivity claims. A second check is the paper's own observation in Figure 1 that the 7 T extrapolated $R_1$ at 5% w/w Miller's LB agar falls below the 4% w/w value: measuring $T_1$ at 7 T across all concentrations would show how often the power-law extrapolation fails.","tokens_in":21142,"feed_emoji":"🧲","tokens_out":12393,"duration_ms":97477,"temperature":0.7,"pith_summary":"This paper aims to turn MRI phantom construction into a data-driven choice rather than a literature hunt. It reports T1, T2, and complex permittivity for seven hydrogel or polymer materials, three paramagnetic salt solutions, three oils, and reference liquids at 0.35, 1.5, and 3 T, with 0.55 T interpolations and 7 T extrapolations. The central result is that generic field-strength rules do not always hold: T1 did not rise with field for every material, r1 did not always fall, and several relaxivities were essentially field-independent. The authors recommend sodium alginate, Miller's LB agar, and xanthan gum as hydrogel candidates, Mn(NO3)2 as the strongest relaxer among the salts, and canola, castor, and grapeseed oils as low-dielectric phantoms.","feed_headline":"0.35, 1.5, 3 T: MRI phantom recipes measured and tabled","feed_subtitle":"The tables give phantom builders field- and concentration-dependent T1, T2, and permittivity values.","key_machinery":"The machinery is the measurement and fitting protocol: inversion-recovery EPI for T1, CPMG multi-spin-echo for T2, and open-ended coaxial-probe permittivity sweeps run through a Cole-Cole relaxation fit to extract the static dielectric constant $\\varepsilon_s$ and static electrical conductivity $\\sigma_s$. Linear fits of relaxation rate $R_n = R_{n,0} + r_n[C]$ versus concentration convert raw relaxation times into relaxivities, and a power law $T_1(B_0) = A\\,B_0^C$ is used to interpolate 0.55 T and extrapolate 7 T values.","core_discovery":"The paper's claim is that these measurements form a reliable, internally consistent dataset for common MRI phantom constituents across clinically relevant field strengths, with concentration sweeps wide enough to extract relaxivities and conductivity slopes. On its own terms, the important findings are that T1 decreased with increasing concentration except for PEG; that the T1 and T2 relaxivities of PEG, PVP, sodium polyacrylate, and deionized water did not vary significantly with field strength; that the T2 relaxivities of gelatin and sodium alginate did not vary significantly; and that the common assumption that T1 rises or r1 falls with field strength fails for several samples. The authors further conclude that sodium alginate, Miller's LB agar, and xanthan gum are good hydrogel candidates, Mn(NO3)2 has the highest relaxivities, and all three oils are good low-dielectric (epsilon_s < 10) phantom materials with low conductivities.","pith_inferences":["Because the dielectric probe is rated only down to 200 MHz, the 14 MHz and static permittivity values rest on model extrapolation; a direct low-frequency measurement would tell whether the reported $\\varepsilon_s$ and $\\sigma_s$ are accurate.","The paper characterizes single or duplex materials only, so combined phantoms (gel plus paramagnetic salt) may behave differently since electrolytes can affect gelling; combined recipes need direct validation before adoption.","If the field independence of PEG, PVP, and sodium polyacrylate relaxivities extends beyond 3 T, those polymers could become transferable phantom standards at ultra-high field, but the 7 T numbers here are extrapolations, not measurements.","A natural test is to build a multi-compartment phantom from the recommended materials and image it at 0.55 T and 7 T; those results would extend or correct the interpolation and extrapolation tables."],"forward_implications":["Concentration-tuning recipes become practical: Mn(NO3)2 changes T1 and T2 at sub-millimolar concentrations, while CuSO4 and NiCl2 need higher concentrations to reach the same relaxation rates.","For PEG, PVP, and sodium polyacrylate, the field-independent relaxivities mean a recipe calibrated at one clinical field should transfer to another without re-measurement.","The oils' low dielectric constants (epsilon_s < 10) and low conductivities make them suitable for high-field phantom compartments where dielectric artifacts must be minimized.","Because conductivity rises linearly with concentration for most hydrogels and salts, RF-heating and SAR phantoms can be tuned by concentration; PEG and PVP are the exceptions.","The 0.55 T and 7 T values are offered as starting points, with the paper itself warning that some extrapolations (for example, 5% w/w Miller's LB agar at 7 T) are not trustworthy."],"supporting_citations":[{"why":"Supplies the power-law relation $T_1(B_0) = A\\,B_0^C$ used to interpolate 0.55 T and extrapolate 7 T relaxation times.","marker":"2"},{"why":"Provides the agar and xanthan gum phantom alternatives whose concentration ranges this study adopts.","marker":"4"},{"why":"Supplies the polyvinyl alcohol preparation method, including the freeze-thaw route, used for the PVA hydrogels.","marker":"5"},{"why":"Provides the polyvinylpyrrolidone concentration calibration for diffusion phantoms that this study extends to relaxation and permittivity measurements.","marker":"8"},{"why":"Gives the 3 T relaxivity results for aqueous paramagnetic solutions used as a comparison for the salt data.","marker":"15"},{"why":"Supplies electrical-property data for vegetable oils that support the use of canola, castor, and grapeseed oils as low-dielectric phantom materials.","marker":"16"},{"why":"Supplies the coaxial-probe permittivity measurement technique used with the dielectric probe.","marker":"20"},{"why":"Defines the Cole-Cole relaxation model used to fit the complex permittivity spectra and extract static dielectric and conductivity.","marker":"22"},{"why":"Provides 0.5 T relaxation-rate values for copper and nickel solutions used as a baseline for the 0.35 T measurements.","marker":"31"}],"fun_headline_variants":["MRI phantom recipes: T1, T2, permittivity at 0.35, 1.5, 3 T","Hydrogel and oil MRI phantoms: field-dependent T1, T2, permittivity","T1, T2, and permittivity mapped for MRI phantom materials at 3 field strengths","New data: T1, T2, and permittivity for MRI phantoms at 0.35–3 T","Best MRI phantoms: sodium alginate, agar, xanthan gum, and oils"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that dielectric-probe readings taken down to 14 MHz, far below the probe's rated 200 MHz lower limit, can be extrapolated through a Cole-Cole fit into accurate static dielectric constants and conductivities; the paper's own Figure 1 note says one 7 T relaxation extrapolation is false, which shows the wider extrapolation chain can break.","fun_headline_variants_meta":{"raw":{"variants":["MRI phantom recipes: T1, T2, permittivity at 0.35, 1.5, 3 T","Hydrogel and oil MRI phantoms: field-dependent T1, T2, permittivity","T1, T2, and permittivity mapped for MRI phantom materials at 3 field strengths","New data: T1, T2, and permittivity for MRI phantoms at 0.35–3 T","Best MRI phantoms: sodium alginate, agar, xanthan gum, and oils"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000259,"raw_usage":{"total_tokens":1631,"prompt_tokens":1035,"completion_tokens":596,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":651,"completion_tokens_details":{"reasoning_tokens":459}},"tokens_in":651,"tokens_out":596,"duration_ms":4668,"temperature":1.0,"reasoning_tokens":459,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:50:57.779822+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same samples with a dielectric probe rated below 14 MHz, or with time-domain reflectometry, and compare the independently obtained $\\varepsilon_s$ and $\\sigma_s$ with the extrapolated values in Tables 3-15; disagreement beyond the reported uncertainties would overturn the permittivity claims. A second check is the paper's own observation in Figure 1 that the 7 T extrapolated $R_1$ at 5% w/w Miller's LB agar falls below the 4% w/w value: measuring $T_1$ at 7 T across all concentrations would show how often the power-law extrapolation fails.","supporting_citations":[{"cited_title":") using: 𝜎(𝜔)= 𝜀଴𝜔𝜀","cited_arxiv_id":null,"evidence_quote":"Supplies the power-law relation $T_1(B_0) = A\\,B_0^C$ used to interpolate 0.55 T and extrapolate 7 T relaxation times."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the agar and xanthan gum phantom alternatives whose concentration ranges this study adopts."},{"cited_title":"We observed T2 decreasing with increasing field strength in PEG, PVA, sodium polyacrylate, and Mn(NO3)2","cited_arxiv_id":null,"evidence_quote":"Supplies the polyvinyl alcohol preparation method, including the freeze-thaw route, used for the PVA hydrogels."},{"cited_title":"Effect of monovalent-divalent cation exchange on the swelling of polyacrylate hydrogels in physiological salt solutions","cited_arxiv_id":null,"evidence_quote":"Provides the polyvinylpyrrolidone concentration calibration for diffusion phantoms that this study extends to relaxation and permittivity measurements."},{"cited_title":"MR relaxation times of agar-based tissue-mimicking phantoms [published online ahead of print 20220412]","cited_arxiv_id":null,"evidence_quote":"Gives the 3 T relaxivity results for aqueous paramagnetic solutions used as a comparison for the salt data."},{"cited_title":"Fabrication of a spherical inclusion phantom for validation of magnetic resonance-based magnetic susceptibility imaging [published online ahead of print 20190805]","cited_arxiv_id":null,"evidence_quote":"Supplies electrical-property data for vegetable oils that support the use of canola, castor, and grapeseed oils as low-dielectric phantom materials."},{"cited_title":"Aqueous paramagnetic solutions for MRI phantoms at 3 T: A detailed study on relaxivities","cited_arxiv_id":null,"evidence_quote":"Supplies the coaxial-probe permittivity measurement technique used with the dielectric probe."},{"cited_title":"A heterogeneous human tissue mimicking phantom for RF heating and MRI thermal monitoring verification [published online ahead of print 20120320]","cited_arxiv_id":null,"evidence_quote":"Defines the Cole-Cole relaxation model used to fit the complex permittivity spectra and extract static dielectric and conductivity."},{"cited_title":"Water in agarose gels studied by nuclear magnetic resonance relaxation in the rotating frame","cited_arxiv_id":null,"evidence_quote":"Provides 0.5 T relaxation-rate values for copper and nickel solutions used as a baseline for the 0.35 T measurements."}],"review_version":1}