Abstract

Ultra low-field MRI expands global access to neuroimaging butproducesscanswithlowsignal-to-noiseratio,reducedcontrast,and thickslices.Whileregression-basedsuper-resolutioncanrecoveranatomicaldetailforsegmentation,itreturnsasingledeterministicestimatethat gives no indication of regions where the low-field input leaves anatomy underdetermined. Generative diffusion models offer an alternative by sampling the posterior distribution of plausible high-field images, quantifyingthisanatomicalambiguity.However,applyingthemto3DwholebrainMRIisrestrictedbymemorybottlenecks,slowsampling,andscannerdomainshifts.Weproposea3Dresidualwaveletdiffusionmodelthat combinesthreeideastoovercomethesehurdles.Alosslesswaveletreparameterisation shrinks the spatial grid to fit a whole brain on a single GPU, residual shifting accelerates sampling by starting from the lowfield input, and domain randomisation promotes scanner generalisation without paired training data. As the high-field reference is not a voxelaligned ground truth, we evaluate downstream volumetric agreement. On a healthy cohort (n = 19) imaged at 0.064T and 3T, our method matchesaleadinggeneral-purposeregressionapproachinvolumetricaccuracy while additionally generating per-voxel uncertainty maps highlighting underdetermined regions. Furthermore, on a pilot dataset (n= 11) of participants with cognitive impairment, disease-relevant atrophy ispreservedratherthannormalisedtowardsahealthyprior.Ourframeworkbringswhole-brainposteriorsamplingtolow-fieldsuper-resolution without sacrificing volumetric accuracy.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/SASHIMI_034.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to Open Review

Open Review Page: https://openreview.net/forum?id=JhR6oNPvoc

BibTex

@InProceedings{YeoRui_UncertaintyAware_MICCAISAT2026,
        author = { Yeow, Rui W. AND Beament, Millie AND Dick, Fred AND Razin, Raha AND Bocchetta, Martina AND Thomas, David L. AND Tregidgo, Henry F. J. AND Alexander, Daniel C. AND Cole, James H.},
        title = { { Uncertainty-Aware 3D Residual Wavelet Diffusion for Ultra Low-Field MRI Super-Resolution } },
        booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
        year = {2026},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 17258},
        month = {pending},
        page = {pending}
}


back to top