arxiv: 2605.01918 · v1 · submitted 2026-05-03 · ⚛️ physics.med-ph

Recognition: 2 theorem links

· Lean Theorem

Characterising Water Exchange in Gliomas Using Diffusion MRI with Free Gradient Waveforms

Arthur Chakwizira, Carl-Fredrik Westin, Filip Szczepankiewicz, Linda Knutsson, Markus Nilsson, Pia C Sundgren

Authors on Pith no claims yet

Pith reviewed 2026-05-08 19:26 UTC · model grok-4.3

classification ⚛️ physics.med-ph

keywords diffusion MRIwater exchangegliomaResExtransmembrane permeabilityfree gradient waveformsbrain tumoraquaporin

0 comments

The pith

Diffusion MRI with free gradient waveforms maps water exchange rates across cell membranes in gliomas.

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

The paper shows that a specialized form of diffusion MRI can measure how quickly water moves in and out of cells inside gliomas without needing invasive procedures. By using free gradient waveforms in a protocol called ResEx, the method separates the effects of water exchange from those of restricted diffusion inside cells. In scans of thirteen patients, the resulting maps displayed uneven exchange patterns, with higher rates often appearing at tumor edges and in some surrounding areas. This information is not available from ordinary diffusion scans and could point to differences in cell size, membrane properties, or water-channel activity that matter for how tumors grow and respond to treatment.

Core claim

The Restriction-Exchange (ResEx) approach with free gradient waveforms permits voxel-wise estimation of water exchange rates in gliomas that is independent of restricted diffusion effects. In thirteen patients the derived maps showed heterogeneous exchange, commonly elevated in enhancing tumor margins and sometimes extending into non-enhancing peritumoural tissue, while oedema values were modestly above those in healthy brain.

What carries the argument

The Restriction-Exchange (ResEx) protocol, which acquires multiple diffusion-weighted images with free gradient waveforms tuned for selective exchange sensitivity to isolate transmembrane water exchange from intracellular restriction.

Load-bearing premise

The ResEx protocol isolates and quantifies water exchange rate without being confounded by restricted diffusion or other tissue properties in the heterogeneous glioma environment.

What would settle it

Direct comparison of ResEx exchange-rate maps with histological measurements of cell density, membrane permeability, and aquaporin-4 expression in the same resected tissue samples showing no systematic correspondence.

Figures

Figures reproduced from arXiv: 2605.01918 by Arthur Chakwizira, Carl-Fredrik Westin, Filip Szczepankiewicz, Linda Knutsson, Markus Nilsson, Pia C Sundgren.

**Figure 1.** Figure 1: shows an overview of the expected effects of the two exchange-encoding waveforms in different types of microstructures. In the presence of restricted diffusion without exchange, the waveforms produce highly similar signal-versus-b curves with negligible contrast. Introducing exchange gives rise to notable contrast between the signals from the two waveforms, with the effect becoming more pronounced at highe… view at source ↗

**Figure 2.** Figure 2: Water exchange estimation in gliomas. Post-Gadolinium T1-weighted images are shown alongside FLAIR in three glioblastoma and two astrocytoma patients. Red ROIs demarcate whole tumour on the T1- weighted images. FLAIR hyperintensity generally colocalises with elevated ADC and reduced kurtosis. Tumours consistently show elevated exchange rates, save for high-ADC, likely fluid-filled regions. Exchange elevati… view at source ↗

**Figure 3.** Figure 3: Comparison of exchange rates between gliomas and contralateral healthy brain tissue in thirteen participants. Leftmost panel shows an example post-Gd T1-weighted image reflecting the tumour (red), healthy white matter (green) and healthy grey matter (yellow). Exchange rates are significantly higher in gliomas compared to both healthy white matter (p = 0.0002) and grey matter (p = 0.0005). The exchange maps… view at source ↗

read the original abstract

Transmembrane water permeability, which regulates cellular water exchange and is influenced by water channels such as aquaporin-4 (AQP4), has been implicated in glioma progression and may affect tumour infiltration and treatment response. Non-invasive mapping of water exchange may therefore provide biomarkers of glioma pathology. This study investigates the feasibility of characterizing water exchange in gliomas using diffusion MRI with free gradient waveforms, known as the Restriction-Exchange (ResEx) approach, which enables exchange quantification independent of restricted diffusion effects. Thirteen patients with histologically confirmed gliomas (ten glioblastomas, three astrocytomas) underwent preoperative MRI at 3T using a custom ResEx protocol. Multiple diffusion-weighted acquisitions with selective exchange sensitivity were performed to estimate voxel-wise maps of the apparent diffusion coefficient (ADC), diffusion kurtosis, and water exchange rate. ResEx-derived maps revealed heterogeneous spatial patterns across and within tumours. Elevated exchange rates were commonly observed in enhancing tumour margins, potentially reflecting smaller cells, increased membrane permeability or AQP4 upregulation. In some cases, elevated exchange extended into non-enhancing peritumoural regions. Exchange values in oedema were slightly higher than in healthy tissue, suggesting potential infiltration or membrane disruption. Diffusion MRI with free gradient waveforms permits non-invasive mapping of water exchange in gliomas and reveals physiological information not captured by standard imaging. Exchange rate mapping may offer novel biomarkers of tumour aggressiveness, infiltration, and treatment response, and holds promise for surgical and radiotherapy planning.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

ResEx gets applied to gliomas in patients and turns up exchange patterns that differ from ADC and kurtosis, but the separation from restriction effects still rests on an untested assumption in this heterogeneous tissue.

read the letter

The main thing to know is that this feasibility study takes the ResEx free-gradient-waveform protocol into glioma patients for the first time and produces voxel-wise exchange maps with higher values in enhancing margins and some extension into non-enhancing peritumoral regions. Those spatial patterns are new relative to the cited prior work and are not explained by standard diffusion metrics alone. They scanned thirteen patients at 3T with a custom protocol, generated the maps, and tied the findings to possible AQP4 effects or infiltration, which is a reasonable clinical hook to explore. The work does a clean job showing that the method can be run in this population and that the resulting contrast looks distinct from what ADC and kurtosis already give. That part is useful for anyone thinking about new diffusion contrasts in neuro-oncology. The soft spot is exactly the one the stress-test note flags. The claim that ResEx cleanly isolates exchange from restricted diffusion across the range of cell sizes and packing densities in gliomas is assumed rather than shown with a direct check. No forward simulations with ground-truth exchange in multi-compartment geometries that match glioma histology, and no paired comparison to an independent method like FEXI, appear in the provided text. Without those, the attributed patterns could still carry some residual restriction leakage, especially in the heterogeneous tumor environment. Model fitting details, error bars, and how the exchange parameter is extracted are also not laid out, which makes it harder to judge how robust the numbers are. The citation pattern looks appropriate and builds on the earlier ResEx papers without obvious gaps. This paper is for diffusion MRI method developers and neuro-oncologists who want to test new biomarkers for infiltration or treatment response. A reader working on advanced waveform designs might pick up ideas from the spatial findings and want to run follow-up experiments. It deserves a serious referee because the patient application is novel and the potential biomarker angle is clear, even if the current evidence is preliminary. I would send it for review and ask the authors to add a direct test of the separation assumption and more quantitative validation details.

Referee Report

2 major / 2 minor

Summary. The manuscript reports a feasibility study applying the Restriction-Exchange (ResEx) diffusion MRI protocol with free gradient waveforms to non-invasively map water exchange rates in 13 patients with histologically confirmed gliomas (10 GBM, 3 astrocytomas). Voxel-wise maps of ADC, diffusion kurtosis, and exchange rate are generated from selective-exchange-sensitive acquisitions at 3T; the maps show heterogeneous intra- and inter-tumoral patterns, with elevated exchange commonly observed at enhancing margins and extending into some non-enhancing peritumoral regions, interpreted as possible signatures of smaller cells, increased permeability, AQP4 upregulation, or infiltration.

Significance. If the ResEx protocol demonstrably decouples transmembrane exchange from intra-compartment restriction across the range of cell sizes and packing densities present in gliomas, the work would supply a new physiological biomarker orthogonal to standard DWI metrics, with potential relevance to tumor aggressiveness, infiltration assessment, and treatment planning. The reported spatial patterns are physiologically plausible and extend beyond conventional imaging contrast.

major comments (2)

[Abstract] Abstract: the central claim that ResEx 'enables exchange quantification independent of restricted diffusion effects' is presented as a property of the acquisition scheme, yet the manuscript provides no forward simulations with known ground-truth exchange in multi-compartment geometries matching GBM histology, nor any paired comparison against an independent exchange method (e.g., FEXI) to verify that recovered exchange rates remain unbiased when restriction parameters vary across tumor subregions.
[Results] Results (voxel-wise maps): no quantitative assessment of fit quality, parameter uncertainty, or residual restriction leakage is reported for the exchange-rate maps; without these, it is impossible to determine whether the observed elevation at enhancing margins reflects true exchange differences or residual confounding from the heterogeneous restriction environment.

minor comments (2)

[Abstract] The abstract would be strengthened by stating the exact number of diffusion-weighted acquisitions, b-values, and waveform parameters used in the custom ResEx protocol.
Clarify whether the reported exchange values are apparent or model-derived parameters and how they are normalized across patients.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive review of our manuscript. We value the recognition of the potential of the ResEx protocol as a new biomarker in gliomas. We address the major comments below and have prepared revisions to the manuscript where the concerns are valid.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim that ResEx 'enables exchange quantification independent of restricted diffusion effects' is presented as a property of the acquisition scheme, yet the manuscript provides no forward simulations with known ground-truth exchange in multi-compartment geometries matching GBM histology, nor any paired comparison against an independent exchange method (e.g., FEXI) to verify that recovered exchange rates remain unbiased when restriction parameters vary across tumor subregions.

Authors: The independence from restricted diffusion is a core design feature of the ResEx method using free gradient waveforms, as described in the foundational methodological papers. This clinical feasibility study does not include new ground-truth simulations or head-to-head comparisons with FEXI. We will revise the abstract to state that the ResEx protocol is designed to enable exchange quantification independent of restriction effects (with appropriate citations), rather than presenting the current patient data as a new demonstration of that independence. revision: partial
Referee: [Results] Results (voxel-wise maps): no quantitative assessment of fit quality, parameter uncertainty, or residual restriction leakage is reported for the exchange-rate maps; without these, it is impossible to determine whether the observed elevation at enhancing margins reflects true exchange differences or residual confounding from the heterogeneous restriction environment.

Authors: We agree that quantitative measures of fit quality and potential residual effects are needed to support interpretation of the exchange maps. In the revised manuscript we will add voxel-wise goodness-of-fit metrics (e.g., R²), parameter uncertainty estimates, and residual analysis to assess whether restriction leakage could contribute to the observed spatial patterns. revision: yes

Circularity Check

0 steps flagged

No significant circularity; exchange-rate maps are data-driven estimates from ResEx acquisitions.

full rationale

The paper presents voxel-wise estimation of water exchange rate directly from custom ResEx diffusion acquisitions in glioma patients. No equations are shown that define the target quantity in terms of itself or that rename a fitted parameter as an independent prediction. The claim that ResEx isolates exchange independent of restriction is stated as a property of the protocol rather than re-derived here; any supporting reference to prior method development is not load-bearing for the present feasibility results in gliomas. The reported spatial patterns and elevated exchange in tumour margins are therefore outputs of the fitting procedure applied to new data, not tautological restatements of the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of the ResEx model in tumor tissue; limited information available from abstract prevents exhaustive listing.

axioms (1)

domain assumption ResEx approach enables exchange quantification independent of restricted diffusion effects
Stated directly in the abstract as the basis for the method's feasibility in gliomas.

pith-pipeline@v0.9.0 · 5583 in / 1043 out tokens · 69948 ms · 2026-05-08T19:26:33.436094+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith.Cost.FunctionalEquation washburn_uniqueness_aczel unclear
ln(S/S0) = −b[E_D + V_ω E_R] + (1/2) b² (3 V_D + 2 V_ω C_{D,R} + V_ω² V_R)·(1 − k Γ)

Reference graph

Works this paper leans on

10 extracted references · 5 canonical work pages

[1]

Centre for Medical Imaging and Function, Skane University Hospital, Lund, Sweden Corresponding author: Arthur Chakwizira Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States Email address: achakwizira@bwh.harvard.edu Sponsors/Grant numbers: VR (Swedish Research Council) o 2024-04968 o 2023-02412 eSSENCE ...

2024
[2]

which has previously demonstrated ability to differentiate between gliomas and meningiomas (Lampinen et al., 2017). However, FEXI is susceptible to biased exchange estimation due to effects of restricted diffusion and imaging gradients (Lasič et al., 2018; Ohene et al., 2023; Ulloa et al., 2017). A more recent diffusion MRI approach using free gradient w...

2017
[3]

(Fig. 1A-B). The waveforms were differentially sensitive to exchange but similarly sensitive to restricted diffusion, that is, they had different Γ (10 and 40 ms) but equal 𝑉# (3500 s-2). Due to scan time constraints, it was not feasible to also include waveforms with variable sensitivity to restrictions. Waveforms were one-dimensional (linear tensor enco...

2023
[4]

+16𝑏%𝐸"%𝐾'⋅(1−𝑘Γ) (5) where 𝐾'=3𝑉

with extrapolated references (Nilsson et al, 2016). The images were then powder-averaged for further analysis. To allow comparison between exchange measurements and contrast enhancement, diffusion-weighted images for each 8 participant were rigidly coregistered with the corresponding T1-weighted volume. FLAIR images were also coregistered to the T1-weight...

2016
[5]

Contralateral healthy grey and white matter were also segmented using SynthSeg (Billot et al., 2023)

with the glioma pre- and post-treatment segmentor (Jain et al., 2025), which delineated three mutually exclusive regions of interest: oedema, tumour core, and contrast-enhancing parts of the tumour. Contralateral healthy grey and white matter were also segmented using SynthSeg (Billot et al., 2023). Quantitative comparisons of exchange estimates in the tu...

2025
[6]

(Fig. 2-3). The variability in exchange effects was not consistently predicted by gadolinium enhancement, suggesting that exchange imaging provides complementary information to conventional contrast-enhanced MRI (Fig. 2). Oedema generally showed low exchange rates relative to enhancing tumour, but slightly elevated rates relative to healthy white matter (...

work page doi:10.1007/s00234-020-02425-9 2016
[7]

https://doi.org/10.1186/2045-8118-10-18 19 Chakwizira, A., Westin, C.-F., Brabec, J., Lasič, S., Knutsson, L., Szczepankiewicz, F., & Nilsson, M. (2022). Diffusion MRI with pulsed and free gradient waveforms: Effects of restricted diffusion and exchange. NMR in Biomedicine, n/a(n/a), e4827. https://doi.org/10.1002/nbm.4827 Chakwizira, A., Zhu, A., Foo, T....

work page doi:10.1186/2045-8118-10-18 2045
[8]

state of the science

https://doi.org/10.1186/s12987-023-00422-7 Omuro, A., & DeAngelis, L. M. (2013). Glioblastoma and other malignant gliomas: A clinical review. JAMA, 310(17), 1842–1850. https://doi.org/10.1001/jama.2013.280319 Ostrom, Q. T., Bauchet, L., Davis, F. G., Deltour, I., Fisher, J. L., Langer, C. E., Pekmezci, M., Schwartzbaum, J. A., Turner, M. C., Walsh, K. M.,...

work page doi:10.1186/s12987-023-00422-7 2013
[9]

https://doi.org/10.3390/cancers12030667 Szczepankiewicz, F., Sjölund, J., Ståhlberg, F., Lätt, J., & Nilsson, M. (2019). Tensor-valued diffusion encoding for diffusional variance decomposition (DIVIDE): Technical feasibility in clinical MRI systems. PLOS ONE, 14(3), e0214238. https://doi.org/10.1371/journal.pone.0214238 Tan, Y., Zhang, H., Zhao, R.-F., Wa...

work page doi:10.3390/cancers12030667 2019
[10]

Denoising of diffusion MRI using random matrix theory

https://doi.org/10.1016/j.neuroimage.2016.08.016 Verkman, A. S., Hara-Chikuma, M., & Papadopoulos, M. C. (2008). Aquaporins—New players in cancer biology. Journal of Molecular Medicine (Berlin, Germany), 86(5), 523–529. https://doi.org/10.1007/s00109-008-0303-9 Wang, Z., Wang, B., Li, Z., Han, G., Meng, C., Jiao, B., Guo, K., Hsu, Y.-C., Sun, Y., Liu, Y.,...

work page doi:10.1016/j.neuroimage.2016.08.016 2016