Pith. sign in

REVIEW 2 major objections 5 minor 79 references

A single closed pore can fake exchange in DEXSY: the rate tracks the first Laplacian eigenvalue π²D/L².

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-13 16:36 UTC pith:UIEXVDJB

load-bearing objection Clean minimal counter-example: localization alone can fake DEXSY/FEXSY exchange rates at k ≈ π²D/L², so the methods are not specific to barrier permeation. the 2 major comments →

arxiv 2603.27907 v1 pith:UIEXVDJB submitted 2026-03-29 cond-mat.mtrl-sci physics.app-phphysics.med-ph

Localization-driven exchange contrast in diffusion exchange spectroscopy

Teddy X Cai , Nathan H Williamson , Peter J Basser This is my paper
classification cond-mat.mtrl-sci physics.app-phphysics.med-ph
keywords diffusion exchange spectroscopyDEXSYFEXSYlocalization regimeedge enhancementrestricted diffusionLaplacian eigenvaluesapparent exchange rate
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Diffusion exchange spectroscopy (DEXSY) is usually read as evidence that water is hopping between compartments of different mobility, often interpreted as membrane permeation in tissue. This paper shows that the same mixing-time contrast can appear inside one sealed one-dimensional compartment with no barriers and no relaxation. Strong gradients localize magnetization near the walls (edge enhancement). During the storage interval that magnetization redistributes by ordinary diffusion; the slowest redistribution rate is the first non-zero eigenvalue of the Laplacian, π²D/L². Fitting the resulting signal decay with the usual first-order exchange model therefore returns an apparent rate k ≈ π²D/L² even though nothing has crossed a membrane. The result is a minimal counter-example: DEXSY and the related method FEXSY are not automatically specific to genuine barrier exchange.

Core claim

In the localization regime ℓ_D/2 ≲ ℓ_g ≲ ℓ_D ≲ L, a single reflecting one-dimensional compartment of length L produces mono-exponential-looking DEXSY contrast with mixing time. The fitted apparent exchange rate is typically k ≈ π² D/L² (more generally ∼ D/L²), which is exactly the slowest non-zero eigenvalue of the diffusion operator. The mechanism is relaxation of the spatial magnetization modes excited by the first gradient encoding; no inter-compartment transfer is required.

What carries the argument

Localization-driven exchange: the first constant-gradient spin-echo encoding leaves a non-uniform magnetization profile that projects onto the cosine eigenmodes of the reflecting Laplacian; subsequent free diffusion during t_m damps those modes at rates λ_n = D(π n/L)², and a phenomenological mono-exponential fit recovers a rate dominated by λ_1.

Load-bearing premise

The three-parameter mono-exponential fit that was designed for two-site barrier exchange still correctly reports the slowest Laplacian eigenvalue when the true process is multi-mode relaxation inside one closed pore.

What would settle it

Simulate or measure DEXSY on a monodisperse set of sealed, non-relaxing pores of known L and D in the stated length-scale window and check whether the fitted k equals π² D/L² within experimental error; a systematic mismatch would refute the claim.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The manuscript shows that a single one-dimensional compartment with reflecting boundaries and no relaxation can produce DEXSY mixing-time contrast in the localization regime. Using a first-order Lie–Trotter matrix evolution of the Bloch–Torrey equation (validated against Monte Carlo), the authors map the window ℓ_D/2 ≲ ℓ_g ≲ ℓ_D ≲ L in which a phenomenological monoexponential fit yields an apparent rate k ≈ π² D/L² (more generally ∼ D/L²), equal to the first non-zero Laplacian eigenvalue. The same rate appears under full 2-D ILT sampling and under FEXSY ADC recovery. The central claim is that DEXSY/FEXSY contrast is therefore not specific to genuine barrier permeation.

Significance. If the result holds, it supplies a clean, minimal counter-example that weakens the common interpretation of DEXSY/FEXSY as uniquely reporting transmembrane exchange. The work is carefully delimited by non-dimensional length-scale ratios, the numerical method is validated and convergent, and the spectral attribution (k tracking λ_1) is falsifiable. Appendices extend the claim to ILT and FEXSY readouts without free parameters tuned to force the equality. This is a useful cautionary result for the porous-media and tissue-microstructure communities and opens a possible route to size estimation via localization-driven mode relaxation.

major comments (2)
  1. The monoexponential model of Eq. (9) is the sole reporter of k, yet the paper itself documents residual multi-exponential bias (Fig. 2 inset) and elevated k when higher modes dominate (Fig. 5, ℓ_D/L, ℓ_g/L ≲ 0.2). A short quantitative bound—e.g., the fractional contribution of n≥2 modes to the fitted k across the homogeneous band of Figs. 3b–c—would make the claimed equality k ≈ π² D/L² fully transparent rather than phenomenological. This is load-bearing for the spectral interpretation but does not overturn the numerical counter-example.
  2. Section 4.2 correctly notes that realistic tissue (surface relaxation, branching, permeability, polydisperse L) will confound the clean λ_1 result, yet the abstract and conclusion still state that DEXSY/FEXSY “may not be specific to genuine barrier permeation.” A single clarifying sentence that the present mechanism is one possible, not necessarily dominant, source of contrast would prevent over-reading of the minimal-system result.
minor comments (5)
  1. Figure 3a color scale and the arbitrary detectability threshold β_1 ≥ 0.02 should be stated in the caption; the threshold is free and affects the reported region.
  2. Eq. (1) waveform and the CGSE ≡ δ=Δ identification are clear, but a one-line reminder that RF pulses are instantaneous would help readers less familiar with the sequence.
  3. Appendix A ILT regularization (λ=10^{-4}) is chosen by a rough L-curve; a brief sensitivity check or reference to the L-curve figure (even if not shown) would strengthen reproducibility.
  4. Typographical: “di ffusion” and similar hyphenation artifacts appear throughout the compiled text; a final proof-read for PDF ligature issues is needed.
  5. The claim that this is the first investigation of localization in double diffusion encoding is plausible but could be softened to “to our knowledge” if any related edge-enhancement work in DDE exists.

Circularity Check

0 steps flagged

No significant circularity: fitted k from independent Bloch-Torrey simulations is compared post-hoc to the known first Laplacian eigenvalue, not defined as or forced by it.

full rationale

The paper generates DEXSY signals by a first-order Lie-Trotter matrix discretization of the Bloch-Torrey equation on a reflecting 1-D interval (Eqs. 2–8), with no free parameters tuned to any target rate. An apparent rate k is then extracted by an independent three-parameter monoexponential fit (Eq. 9) that was originally motivated by two-site exchange but is applied here purely phenomenologically. Across the localization band ℓ_D/2 ≲ ℓ_g ≲ ℓ_D ≲ L the fitted values collapse numerically to ≈ π² D/L² (Figs. 3–4); the authors subsequently note that this number coincides with the first non-zero Neumann eigenvalue of the Laplacian and interpret the match via eigen-decomposition of the magnetization profile (Eqs. 16–20, Fig. 5). The equality is therefore an a-posteriori observation, not a definitional identity or a prediction forced by a fitted input. Self-citations supply only the experimental context and the matrix formalism itself; they do not underwrite the new localization result. Appendices A–B confirm the same rate appears under ILT and FEXSY readouts, again without circular forcing. The derivation chain is self-contained numerical counter-example work.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The central claim rests on standard continuum diffusion NMR, a minimal geometric idealization, and a conventional phenomenological fit. No new physical entities are postulated; free parameters are only numerical discretization choices and the three-parameter exponential model used to extract k.

free parameters (3)
  • phenomenological fit coefficients (β1, k, β3)
    Three free parameters of Eq. (9) are fitted to each simulated S(t_m) curve; k is the reported exchange rate. The functional form is assumed, not derived from the multi-mode spectrum.
  • spatial/temporal discretization (Δx, Δt, p≈0.1)
    Chosen by hand to keep p≪0.5 and Δx≪ℓ_g; convergence is checked but the values remain free numerical parameters of the solver.
  • detectability threshold β1≥0.02
    Arbitrary SNR-based cutoff used to filter the (ℓ_D, ℓ_g) map; does not alter the k≈π²D/L² result inside the retained region but defines the reported regime boundary.
axioms (4)
  • domain assumption Magnetization evolves according to the Bloch–Torrey equation with constant diffusivity D and no relaxation (R1=R2=ρ=0).
    Stated in §1–2; allows isolation of pure localization-driven contrast.
  • domain assumption Domain is a finite 1D interval with Neumann (reflecting) boundaries.
    Minimal counter-example geometry introduced in the introduction and used throughout the numerics.
  • standard math First-order Lie–Trotter splitting of diffusion and phase operators on a discrete grid converges to the continuum solution when p≪0.5 and Δx≪ℓ_g.
    Justified via the product formula and validated against Monte Carlo in §2–3.
  • ad hoc to paper The monoexponential model of barrier-limited exchange remains an adequate phenomenological descriptor of multi-mode decay for extracting an apparent k.
    Eq. (9) is imported from two-site exchange literature and applied without re-derivation to the closed-pore signals.

pith-pipeline@v1.1.0-grok45 · 28966 in / 2856 out tokens · 25157 ms · 2026-07-13T16:36:16.670536+00:00 · methodology

0 comments
read the original abstract

Diffusion exchange spectroscopy (DEXSY) is a method to probe exchange between domains of varying confinement. Analyses of DEXSY signals typically assume Gaussian diffusion within distinct compartments and first-order exchange kinetics between them. Other situations can yield DEXSY signal contrast with respect to mixing time, however, leading to potentially erroneous interpretation. Here, we demonstrate that a one-dimensional compartment with reflecting boundaries and without relaxation can by itself produce such contrast in certain experimental regimes. The origin of this contrast is the diffusive mixing of spin isochromats initially near versus far from either boundary, as the former can be relatively coherent in an effect known as edge enhancement or signal localization. We consider DEXSY signals in the case of extended field gradients and identical encodings. Signals were generated via a numerical approach that solves the Bloch-Torrey equation in discrete space and time using matrix operators. We find that in the localization regime, an apparent first-order rate constant of exchange, $k$, can be extracted from DEXSY signals even in this minimal system. The measured $k$ is approximately proportional to $D/L^2$, where $D$ is the diffusivity and $L$ is the domain size. Typically, $k \approx \pi^2 D/L^2$. We attribute this localization-driven exchange to the relaxation of spatial magnetization modes with mixing time, noting that $\pi^2 D/L^2$ is the first non-zero eigenvalue of the Laplacian basis. These results demonstrate that DEXSY and related methods such as filter exchange spectroscopy (FEXSY) may not be specific to genuine barrier permeation.

Figures

Figures reproduced from arXiv: 2603.27907 by Nathan H Williamson, Peter J Basser, Teddy X Cai.

Figure 1
Figure 1. Figure 1: Absolute magnetization vectors |m| plotted vs. bin midpoints x¯ (solid lines). Parameters were L = 20 µm, T = 10 ms, D = 2 µm2 /ms, with varying g = [0.3, 0.4, 0.6] T/m (light green to dark blue, respectively), and γ ≈ 2.675 × 108 rad/s/T. Discretization was ∆x = 0.2 µm, ∆t = 2 µs and p = 0.1. Initial condition was m(0) = 1. For g = 0.4 T/m, MC simulated data is included for comparison, plotted at every ot… view at source ↗
Figure 2
Figure 2. Figure 2: DEXSY signal S (tm) for the same parameters as in [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Fits of Eq. (9) for signals generated at [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Fitted k for fixed ratios of ℓD/L = 0.5 and ℓg/L = 0.3, and L spaced linearly by 1 from 2−20 µm and D by 0.1 from 0.5−3 µm2 /ms. Discretization was ∆t = 2 µs with ∆x adjusted to maintain p ≈ 0.1. (a) Values of k on a log color scale. Dashed lines indicate 1-D cross-sections shown in the next part. (b) 1-D cross sections of (a) at fixed D = 2 µm2 /ms (left) and L = 10 µm (right). The plot vs. L is on a log-… view at source ↗
Figure 5
Figure 5. Figure 5: Examples of eigen-decomposition of m(T). The ratio ℓg/ℓD = 0.6 was fixed, while ℓD = [0.1, 0.2, 0.3]L was varied (light to dark, respectively). (a) Absolute profiles |m(T)|. Note that fixing ℓg/ℓD yields profiles with similar maxima as bD ∝ (ℓD/ℓg) 6 . (b) Absolute eigen-decomposition coefficients cn in the basis un, plotted on a log y-axis. Note the difference in tails, with increasing high-frequency cont… view at source ↗
Figure 2
Figure 2. Figure 2: The filter bf ≈ 4.6 ms/µm2 corresponds to g = 0.06 T/m, while b2 ≈ [0.13, 0.52] ms/µm2 correspond to g = [0.01, 0.02] T/m (light green and dark blue, respectively). (a) Raw signals S (tm). A fit of Eq. (9) is also shown (dashed line), yielding k ≈ 50 s−1 in both cases. (b) Corresponding Dapp(tm) calculated by Eq. (B.1). A fit of Eq. (9) yields a similar k ≈ 48 s−1 . Note that since this is a recovery proce… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

79 extracted references · 31 canonical work pages

  1. [1]

    Benjamini, P

    D. Benjamini, P . J. Basser, Multidimensional correlati on mri, NMR in Biomedicine 33 (12) (2020). doi:10.1002 /nbm.4226

    2020

  2. [2]

    Topgaard, Multidimensional di ffusion mri, Jour- nal of Magnetic Resonance 275 (2017) 98–113

    D. Topgaard, Multidimensional di ffusion mri, Jour- nal of Magnetic Resonance 275 (2017) 98–113. doi:10.1016/j.jmr.2016.12.007

    work page doi:10.1016/j.jmr.2016.12.007 2017

  3. [3]

    R. N. Henriques, M. Palombo, S. N. Jespersen, N. Shemesh, H. Lundell, A. Ianu¸ s, Double di ffusion encoding and applica- tions for biomedical imaging, Journal of Neuroscience Meth ods 348 (2021) 108989. doi:10.1016 /j.jneumeth.2020.108989

    arXiv 2021

  4. [4]

    P . T. Callaghan, I. Furó, Di ffusion-diffusion correlation and exchange as a signature for local order and dynamics, The Journal of Chemical Physics 120 (8) (2004) 4032–4038. doi:10.1063/1.1642604

    work page doi:10.1063/1.1642604 2004

  5. [5]

    Y . Qiao, P . Galvosas, T. Adalsteinsson, M. Schönho ff, P . T. Callaghan, Di ffusion exchange nmr spectroscopic study 11 of dextran exchange through polyelectrolyte multilayer ca p- sules, The Journal of Chemical Physics 122 (21) (2005). doi:10.1063/1.1924707

    work page doi:10.1063/1.1924707 2005

  6. [6]

    Åslund, A

    I. Åslund, A. Nowacka, M. Nilsson, D. Topgaard, Filter- exchange PGSE NMR determination of cell membrane perme- ability, Journal of Magnetic Resonance 200 (2) (2009) 291–2 95

    2009

  7. [7]

    Lasi ˇc, M

    S. Lasi ˇc, M. Nilsson, J. Lätt, F. Ståhlberg, D. Topgaard, Apparent exchange rate mapping with di ffusion mri, Magnetic Resonance in Medicine 66 (2) (2011) 356–365. doi:10.1002 /mrm.22782

    2011

  8. [8]

    Nilsson, J

    M. Nilsson, J. Lätt, D. van Westen, S. Brockstedt, S. Lasi ˇc, F. Ståhlberg, D. Topgaard, Noninvasive mapping of water di ffu- sional exchange in the human brain using filter-exchange ima g- ing, Magnetic Resonance in Medicine 69 (6) (2013) 1572–1580

    2013

  9. [9]

    Lampinen, F

    B. Lampinen, F. Szczepankiewicz, D. van Westen, E. Englu nd, P . C Sundgren, J. Lätt, F. Ståhlberg, M. Nilsson, Optimal ex- perimental design for filter exchange imaging: Apparent ex- change rate measurements in the healthy brain and in intracr a- nial tumors, Magnetic Resonance in Medicine 77 (3) (2016) 1104–1114. doi:10.1002 /mrm.26195

    2016

  10. [10]

    Benjamini, M

    D. Benjamini, M. E. Komlosh, P . J. Basser, Imaging local di ffusive dynamics using di ffusion exchange spec- troscopy mri, Physical Review Letters 118 (15) (2017). doi:10.1103/physrevlett.118.158003

    work page doi:10.1103/physrevlett.118.158003 2017

  11. [11]

    T. X. Cai, D. Benjamini, M. E. Komlosh, P . J. Basser, N. H. Williamson, Rapid detection of the presence of di ffusion ex- change, Journal of Magnetic Resonance 297 (2018) 17–22. doi:10.1016/j.jmr.2018.10.004

    work page doi:10.1016/j.jmr.2018.10.004 2018

  12. [12]

    N. H. Williamson, R. Ravin, T. X. Cai, D. Benjamini, M. Falgairolle, M. J. O’Donovan, P . J. Basser, Real-time measurement of di ffusion exchange rate in biological tis- sue, Journal of Magnetic Resonance 317 (2020) 106782. doi:10.1016/j.jmr.2020.106782

    work page doi:10.1016/j.jmr.2020.106782 2020

  13. [13]

    T. X. Cai, N. H. Williamson, R. Ravin, P . J. Basser, Disen tan- gling the e ffects of restriction and exchange with di ffusion ex- change spectroscopy, Frontiers in Physics 10 (2022) 805793 . doi:10.3389/fphy.2022.805793

    work page doi:10.3389/fphy.2022.805793 2022

  14. [14]

    T. X. Cai, N. H. Williamson, R. Ravin, P . J. Basser, The di f- fusion exchange ratio (dexr): A minimal sampling of di ffusion exchange spectroscopy to probe exchange, restriction, and time- dependence, Journal of Magnetic Resonance 366 (2024) 107745. doi:https://doi.org/10.1016/j.jmr.2024.107745

    work page doi:10.1016/j.jmr.2024.107745 2024

  15. [15]

    Ramadan, Di ffusion-exchange weighted imaging, Magnetic Resonance Insights 3 (2009)

    S. Ramadan, Di ffusion-exchange weighted imaging, Magnetic Resonance Insights 3 (2009). doi:10.4137 /mri.s3504

    2009

  16. [16]

    R. Bai, Z. Li, C. Sun, Y .-C. Hsu, H. Liang, P . Basser, Feas i- bility of filter-exchange imaging (fexi) in measuring di fferent exchange processes in human brain, NeuroImage 219 (2020) 117039. doi:10.1016 /j.neuroimage.2020.117039

    arXiv 2020

  17. [17]

    J. O. Breen-Norris, B. Siow, C. Walsh, B. Hipwell, I. Hil l, T. Roberts, M. G. Hall, M. F. Lythgoe, A. Ianus, D. C. Alexan- der, S. Walker-Samuel, Measuring diffusion exchange across the cell membrane with dexsy (di ffusion exchange spectroscopy), Magnetic Resonance in Medicine 84 (3) (2020) 1543–1551. doi:10.1002/mrm.28207

    work page doi:10.1002/mrm.28207 2020

  18. [18]

    N. H. Williamson, R. Ravin, T. X. Cai, M. Falgairolle, M. J. O’Donovan, P . J. Basser, Water exchange rates measure activ e transport and homeostasis in neural tissue, PNAS Nexus 2 (3) (2023) pgad056. doi:10.1093 /pnasnexus/pgad056

    2023

  19. [19]

    Z. Li, C. Liang, Q. He, T. Feiweier, Y . Hsu, J. Li, R. Bai, Compar- ison of water exchange measurements between filter-exchang e imaging and di ffusion time-dependent kurtosis imaging in the human brain, Magnetic Resonance in Medicine 93 (6) (2025) 2357–2369. doi:10.1002 /mrm.30454

    2025

  20. [20]

    Kärger, Zur bestimmung der di ffusion in einem zweibere- ichsystem mit hilfe von gepulsten feldgradienten, Annalen der Physik 479 (1–2) (1969) 1–4

    J. Kärger, Zur bestimmung der di ffusion in einem zweibere- ichsystem mit hilfe von gepulsten feldgradienten, Annalen der Physik 479 (1–2) (1969) 1–4. doi:10.1002 /andp.19694790102

    1969

  21. [21]

    Kärger, Nmr self-di ffusion studies in heterogeneous systems, Advances in Colloid and Interface Science 23 (1985) 129–148

    J. Kärger, Nmr self-di ffusion studies in heterogeneous systems, Advances in Colloid and Interface Science 23 (1985) 129–148 . doi:10.1016/0001-8686(85)80018-x

    work page doi:10.1016/0001-8686(85)80018-x 1985

  22. [22]

    I. O. Jelescu, A. de Skowronski, F. Ge ffroy, M. Palombo, D. S. Novikov, Neurite exchange imaging (nexi): A minimal model of di ffusion in gray matter with inter- compartment water exchange, NeuroImage 256 (2022) 119277. doi:10.1016/j.neuroimage.2022.119277

    work page doi:10.1016/j.neuroimage.2022.119277 2022

  23. [23]

    J. H. Jensen, Di ffusional kurtosis time dependence and the water exchange rate for the multi-compartment kärger model , Magnetic Resonance in Medicine 91 (3) (2023) 1122–1135. doi:10.1002/mrm.29926

    work page doi:10.1002/mrm.29926 2023

  24. [24]

    Aird-Rossiter, H

    C. Aird-Rossiter, H. Zhang, D. C. Alexander, D. K. Jones , M. Palombo, Decoding gray matter, large-scale analysis of brain cell morphometry to inform microstructural modeling of diffusion mr signals, Communications Biology 9 (1) (2026). doi:10.1038/s42003-025-09353-5

    work page doi:10.1038/s42003-025-09353-5 2026

  25. [25]

    Khateri, M

    M. Khateri, M. Reisert, A. Sierra, J. Tohka, V . G. Kisele v, What does fexi measure?, NMR in Biomedicine 35 (12) (2022). doi:10.1002/nbm.4804

    work page doi:10.1002/nbm.4804 2022

  26. [26]

    Chakwizira, K

    A. Chakwizira, K. ¸ Sim¸ sek, F. Szczepankiewicz, M. Pal ombo, M. Nilsson, The role of dendritic spines in water exchange me a- surements with di ffusion mri: Double di ffusion encoding and free-waveform mri (2025). doi:10.48550 /ARXIV .2504.21537

    Pith/arXiv arXiv 2025

  27. [27]

    V . G. Kiselev, J.-R. Li, What does fexi measure in neuron s? (2026). doi:10.48550 /ARXIV .2601.20657

    arXiv 2026

  28. [28]

    W. B. Hyslop, P . C. Lauterbur, E ffects of restricted di ffusion on microscopic nmr imaging, Journal of Magnetic Resonance 94 ( 3) (1991) 501–510. doi:10.1016 /0022-2364(91)90136-h

    1991

  29. [29]

    B. Pütz, D. Barsky, K. Schulten, Edge enhancement by dif - fusion in microscopic magnetic resonance imaging, Journal of Magnetic Resonance 97 (1) (1992) 27–53. doi:10.1016 /0022- 2364(92)90235-y

    1992

  30. [30]

    Callaghan, A

    P . Callaghan, A. Coy, L. Forde, C. Rofe, Di ffusive re- laxation and edge enhancement in nmr microscopy, Journal of Magnetic Resonance, Series A 101 (3) (1993) 347–350. doi:10.1006/jmra.1993.1057

    work page doi:10.1006/jmra.1993.1057 1993

  31. [31]

    Stepišnik, A

    J. Stepišnik, A. Duh, A. Mohori ˇc, I. Serša, Mri edge en- hancement as a di ffusive discord of spin phase structure, Journal of Magnetic Resonance 137 (1) (1999) 154–160. doi:10.1006/jmre.1998.1678. 12

    work page doi:10.1006/jmre.1998.1678 1999

  32. [32]

    Özarslan, U

    E. Özarslan, U. Nevo, P . J. Basser, Anisotropy induced by macroscopic boundaries: Surface-normal mapping using diffusion-weighted imaging, Biophysical Journal 94 (7) (2008) 2809–2818. doi:10.1529 /biophysj.107.124081

    2008

  33. [33]

    S. D. Stoller, W. Happer, F. J. Dyson, Transverse spin re laxation in inhomogeneous magnetic fields, Physical Review A 44 (11) (1991) 7459–7477. doi:10.1103 /physreva.44.7459

    1991

  34. [34]

    T. M. de Swiet, P . N. Sen, Decay of nuclear magnetiza- tion by bounded di ffusion in a constant field gradient, The Journal of Chemical Physics 100 (8) (1994) 5597–5604. doi:10.1063/1.467127

    work page doi:10.1063/1.467127 1994

  35. [35]

    Deswiet, Di ffusive edge enhancement in imaging, Jour- nal of Magnetic Resonance, Series B 109 (1) (1995) 12–18

    T. Deswiet, Di ffusive edge enhancement in imaging, Jour- nal of Magnetic Resonance, Series B 109 (1) (1995) 12–18. doi:10.1006/jmrb.1995.1141

    work page doi:10.1006/jmrb.1995.1141 1995

  36. [36]

    M. D. Hürlimann, K. G. Helmer, T. M. de Swiet, P . N. Sen, Sp in echoes in a constant gradient and in the presence of simple re - striction, Journal of Magnetic Resonance, Series A 113 (199 5) 260–264. doi:10.1006 /jmra.1995.1091

    arXiv 1995

  37. [37]

    A. F. Frøhlich, L. Østergaard, V . G. Kiselev, E ffect of impermeable boundaries on di ffusion-attenuated mr signal, Journal of Magnetic Resonance 179 (2) (2006) 223–233. doi:10.1016/j.jmr.2005.12.005

    work page doi:10.1016/j.jmr.2005.12.005 2006

  38. [38]

    D. S. Grebenkov, Di ffusion mri /nmr at high gradients: Chal- lenges and perspectives, Microporous and Mesoporous Materials 269 (2018) 79–82. doi:10.1016 /j.micromeso.2017.02.002

    2018

  39. [39]

    Moutal, K

    N. Moutal, K. Demberg, D. S. Grebenkov, T. A. Kuder, Localization regime in di ffusion nmr: Theory and experi- ments, Journal of Magnetic Resonance 305 (2019) 162–174. doi:10.1016/j.jmr.2019.06.016

    work page doi:10.1016/j.jmr.2019.06.016 2019

  40. [40]

    H. Y . Carr, E. M. Purcell, E ffects of di ffusion on free preces- sion in nuclear magnetic resonance experiments, Physical R e- view 94 (3) (1954) 630–638. doi:10.1103 /physrev.94.630

    1954

  41. [41]

    E. O. Stejskal, J. E. Tanner, Spin di ffusion measurements: Spin echoes in the presence of a time-dependent field gradi- ent, The Journal of Chemical Physics 42 (1) (1965) 288–292. doi:10.1063/1.1695690

    doi:10.1063/1.1695690 1965

  42. [42]

    N. H. Williamson, R. Ravin, D. Benjamini, H. Merkle, M. F al- gairolle, M. J. O’Donovan, D. Blivis, D. Ide, T. X. Cai, N. S. Ghorashi, R. Bai, P . J. Basser, Magnetic resonance measurements of cellular and sub-cellular membrane structures in live an d fixed neural tissue, eLife 8 (2019). doi:10.7554 /elife.51101

    2019

  43. [43]

    N. H. Williamson, R. Ravin, T. X. Cai, J. A. Rey, P . J. Bass er, Passive water exchange between multiple sites can explain w hy apparent exchange rate constants depend on ionic and osmoti c conditions in gray matter, Magnetic Resonance Letters (202 5) 200225doi:10.1016/j.mrl.2025.200225

    work page doi:10.1016/j.mrl.2025.200225 2025

  44. [44]

    Cheng, S

    Z. Cheng, S. Hu, G. Han, K. Fang, X. Jin, A. Ordinola, E. Özarslan, R. Bai, Using deep learning to accelerate magne tic resonance measurements of molecular exchange, The Journal of Chemical Physics 159 (5) (2023). doi:10.1063 /5.0159343

    2023

  45. [45]

    Herberthson, P

    M. Herberthson, P . J. Basser, E. Özarslan, Time-depend ent dif- fusion in one-dimensional disordered media decorated by pe r- meable membranes: Theoretical findings backed by simula- tions and a new disorder class, Physics of Fluids 37 (6) (2025 ). doi:10.1063/5.0272370

    work page doi:10.1063/5.0272370 2025

  46. [46]

    T. X. Cai, N. H. Williamson, R. Ravin, M. Herberthson, E. Özarslan, P . J. Basser, Measuring the velocity autocorre lation function using di ffusion nmr, The Journal of Chemical Physics 162 (17) (2025). doi:10.1063 /5.0258081

    2025

  47. [47]

    P . T. Callaghan, A simple matrix formalism for spin echo anal- ysis of restricted di ffusion under generalized gradient wave- forms, Journal of Magnetic Resonance 129 (1) (1997) 74–84. doi:10.1006/jmre.1997.1233

    work page doi:10.1006/jmre.1997.1233 1997

  48. [48]

    Caprihan, L

    A. Caprihan, L. Wang, E. Fukushima, A multiple-narrow- pulse approximation for restricted diffusion in a time-varying field gra- dient, Journal of Magnetic Resonance, Series A 118 (1) (1996 ) 94–102. doi:10.1006 /jmra.1996.0013

    arXiv 1996

  49. [49]

    Barzykin, Theory of spin echo in restricted geometri es under a step-wise gradient pulse sequence, Journal of Magnetic Re so- nance 139 (2) (1999) 342–353

    A. Barzykin, Theory of spin echo in restricted geometri es under a step-wise gradient pulse sequence, Journal of Magnetic Re so- nance 139 (2) (1999) 342–353. doi:10.1006 /jmre.1999.1778

    arXiv 1999

  50. [50]

    D. S. Grebenkov, Nmr survey of reflected brownian mo- tion, Reviews of Modern Physics 79 (3) (2007) 1077–1137. doi:10.1103/revmodphys.79.1077

    doi:10.1103/revmodphys.79.1077 2007

  51. [51]

    D. S. Grebenkov, Laplacian eigenfunctions in nmr. i. a n umerical tool, Concepts in Magnetic Resonance Part A 32A (4) (2008) 277–301. doi:10.1002 /cmr.a.20117

    2008

  52. [52]

    Herberthson, E

    M. Herberthson, E. Özarslan, H. Knutsson, C.-F. Westin , Dy- namics of local magnetization in the eigenbasis of the bloch - torrey operator, The Journal of Chemical Physics 146 (12) (2017). doi:10.1063 /1.4978621

    2017

  53. [53]

    G. P . Zientara, J. H. Freed, Spin-echoes for di ffusion in bounded, heterogeneous media: A numerical study, The Journal of Chem - ical Physics 72 (2) (1980) 1285–1292. doi:10.1063 /1.439190

    1980

  54. [54]

    M. Blees, The e ffect of finite duration of gradient pulses on the pulsed-field-gradient nmr method for studying restricted d iffu- sion, Journal of Magnetic Resonance, Series A 109 (2) (1994) 203–209. doi:10.1006 /jmra.1994.1156

    arXiv 1994

  55. [55]

    K. M. Salikhov, D. J. Schneider, S. Saxena, J. H. Freed, A the- oretical approach to the analysis of arbitrary pulses in mag netic resonance, Chemical Physics Letters 262 (1–2) (1996) 17–26 . doi:10.1016/0009-2614(96)01044-5

    work page doi:10.1016/0009-2614(96)01044-5 1996

  56. [56]

    P . N. Sen, A. André, S. Axelrod, Spin echoes of nuclear ma gne- tization di ffusing in a constant magnetic field gradient and in a restricted geometry, The Journal of Chemical Physics 111 (1 4) (1999) 6548–6555. doi:10.1063 /1.480009

    1999

  57. [57]

    Crank, P

    J. Crank, P . Nicolson, A practical method for numeri- cal evaluation of solutions of partial di fferential equations of the heat-conduction type, Mathematical Proceedings of the Cambridge Philosophical Society 43 (1) (1947) 50–67. doi:10.1017/s0305004100023197

    doi:10.1017/s0305004100023197 1947

  58. [58]

    H. F. Trotter, On the product of semi-groups of operator s, Pro- ceedings of the American Mathematical Society 10 (4) (1959) 545–551. doi:10.1090 /s0002-9939-1959-0108732-6. 13

    1959

  59. [59]

    C. H. Neuman, Spin echo of spins di ffusing in a bounded medium, The Journal of Chemical Physics 60 (11) (1974) 4508–4511. doi:10.1063 /1.1680931

    1974

  60. [60]

    Le Bihan, E

    D. Le Bihan, E. Breton, D. Lallemand, P . Grenier, E. Ca- banis, M. Laval-Jeantet, Mr imaging of intravoxel inco- herent motions: application to di ffusion and perfusion in neurologic disorders., Radiology 161 (2) (1986) 401–407. doi:10.1148/radiology.161.2.3763909

    work page doi:10.1148/radiology.161.2.3763909 1986

  61. [61]

    Casanova, J

    F. Casanova, J. Perlo, B. Blumich (Eds.), Single-Sided NMR, 2011th Edition, Springer, Berlin, Germany, 2011

    2011

  62. [62]

    Keller, C

    D. Keller, C. Erö, H. Markram, Cell densities in the mous e brain: A systematic review, Frontiers in Neuroanatomy 12 (2018). doi:10.3389/fnana.2018.00083

    doi:10.3389/fnana.2018.00083 2018

  63. [63]

    Ianu¸ s, J

    A. Ianu¸ s, J. Carvalho, F. F. Fernandes, R. Cruz, C. Chav ar- rias, M. Palombo, N. Shemesh, Soma and neurite den- sity mri (sandi) of the in-vivo mouse brain and comparison with the allen brain atlas, NeuroImage 254 (2022) 119135. doi:10.1016/j.neuroimage.2022.119135

    work page doi:10.1016/j.neuroimage.2022.119135 2022

  64. [64]

    Shapson-Coe, M

    A. Shapson-Coe, M. Januszewski, D. R. Berger, A. Pope, Y . Wu, T. Blakely, R. L. Schalek, P . H. Li, S. Wang, J. Maitin-Shepar d, N. Karlupia, S. Dorkenwald, E. Sjostedt, L. Leavitt, D. Lee, J. Troidl, F. Collman, L. Bailey, A. Fitzmaurice, R. Kar, B. Field, H. Wu, J. Wagner-Carena, D. Aley, J. Lau, Z. Lin, D. Wei, H. Pfister, A. Peleg, V . Jain, J. W. Lic...

    2024

  65. [65]

    K. R. Brownstein, C. E. Tarr, Importance of classical di ffusion in nmr studies of water in biological cells, Physical Review A 19 (6) (1979) 2446–2453. doi:10.1103 /physreva.19.2446

    1979

  66. [66]

    Afrough, Magnetic resonance relaxation in heteroge - neous materials is analogous to first-order chemical reac- tion, Transport in Porous Media 151 (7) (2024) 1493–1509

    A. Afrough, Magnetic resonance relaxation in heteroge - neous materials is analogous to first-order chemical reac- tion, Transport in Porous Media 151 (7) (2024) 1493–1509. doi:10.1007/s11242-024-02075-y

    work page doi:10.1007/s11242-024-02075-y 2024

  67. [67]

    D. S. Grebenkov, Exploring di ffusion across permeable barriers at high gradients. ii. localization regime, Journal of Magn etic Resonance 248 (2014) 164–176. doi:10.1016 /j.jmr.2014.08.016

    2014

  68. [68]

    J. H. Lee, C. Labadie, C. S. Springer, G. S. Harbison, Two - dimensional inverse laplace transform nmr: altered relax- ation times allow detection of exchange correlation, Journ al of the American Chemical Society 115 (17) (1993) 7761–7764. doi:10.1021/ja00070a022

    work page doi:10.1021/ja00070a022 1993

  69. [69]

    K. E. Washburn, P . T. Callaghan, Tracking pore to pore exchange using relaxation exchange spectroscopy, Physical Review Letters 97 (17) (2006). doi:10.1103 /physrevlett.97.175502

    2006

  70. [70]

    Ordinola, E

    A. Ordinola, E. Özarslan, R. Bai, M. Herberthson, Limit ations and generalizations of the first order kinetics reaction exp ression for modeling diffusion-driven exchange: Implications on nmr ex- change measurements, The Journal of Chemical Physics 160 (8 ) (2024). doi:10.1063 /5.0188865

    2024

  71. [71]

    H.-H. Lee, N. Wang, E. Fieremans, S. Y . Huang, D. S. Novik ov, Localization regime of di ffusion (lord) in mouse and pig corti- cal gray matter and its sensitivity to soma size, in: Proceed ings of the 31st Annual Meeting of the International Society of Ma g- netic Resonance in Medicine (ISMRM), Toronto, Canada, 2023 , p. 0681

    2023

  72. [72]

    Kac, Can one hear the shape of a drum?, The American Mat h- ematical Monthly 73 (4) (1966) 1

    M. Kac, Can one hear the shape of a drum?, The American Mat h- ematical Monthly 73 (4) (1966) 1. doi:10.2307 /2313748

    1966

  73. [73]

    F. B. Laun, T. A. Kuder, W. Semmler, B. Stieltjes, Determ i- nation of the defining boundary in nuclear magnetic resonanc e diffusion experiments, Physical Review Letters 107 (4) (2011). doi:10.1103/physrevlett.107.048102

    work page doi:10.1103/physrevlett.107.048102 2011

  74. [74]

    Song, Detection of the high eigenmodes of spin di ffu- sion in porous media, Physical Review Letters 85 (18) (2000) 3878–3881

    Y .-Q. Song, Detection of the high eigenmodes of spin di ffu- sion in porous media, Physical Review Letters 85 (18) (2000) 3878–3881. doi:10.1103 /physrevlett.85.3878

    2000

  75. [75]

    P . T. Callaghan, J. Stepišnik, Frequency-domain analy sis of spin motion using modulated-gradient nmr, Journal of Magnetic Resonance, Series A 117 (1) (1995) 118–122. doi:10.1006/jmra.1995.9959

    work page doi:10.1006/jmra.1995.9959 1995

  76. [76]

    E. C. Parsons, M. D. Does, J. C. Gore, Temporal di ffusion spec- troscopy: Theory and implementation in restricted systems us- ing oscillating gradients, Magnetic Resonance in Medicine 55 (1) (2005) 75–84. doi:10.1002 /mrm.20732

    2005

  77. [77]

    Ordinola, S

    A. Ordinola, S. Cai, P . Lundberg, R. Bai, E. Özarslan, On the sampling strategies and models for measuring diffusion exchange with a double diffusion encoding sequence, Magnetic Resonance Letters 3 (3) (2023) 232–247. doi:10.1016 /j.mrl.2023.05.003

    2023

  78. [78]

    Hansen, The l-curve and its use in the numerical treat ment of inverse problems, in: InviteComputational Inverse Proble ms in Electrocardiology, WIT Press, 2000, pp

    P . Hansen, The l-curve and its use in the numerical treat ment of inverse problems, in: InviteComputational Inverse Proble ms in Electrocardiology, WIT Press, 2000, pp. 119–142

    2000

  79. [79]

    Song, Y .-Q

    R. Song, Y .-Q. Song, M. V embusubramanian, J. L. Paulsen , The robust identification of exchange from t2–t2 time-domai n features, Journal of Magnetic Resonance 265 (2016) 164–171 . doi:10.1016/j.jmr.2016.02.001. 14

    work page doi:10.1016/j.jmr.2016.02.001 2016