Morphometric Brain Networks at Ultra-Low Field: A First Assessment of Feasibility and Reliability

Abstract

BackgroundUltra-low field (ULF) magnetic resonance imaging (MRI) offers accessible point-of-care neuroimag-ing, yet its capability to support multivariate network analyses remains understudied.

MethodsWe systematically evaluated the feasibility and reliability of Morphometric Similarity Networks (MSN) and Morphometric Inverse Divergence (MIND) networks derived from paired high-field (HF, 3T/1.5T) and ULF (64mT) scans in a primary cohort (N=23) and an independent replication cohort (N=19). We tested the impact of super-resolution processing, cortical parcellation granularity (Desikan-Killiany vs. Destrieux), and mesh resolution (fsnative vs. fsaverage5) on edge-level cross-scanner agreement (Lins concordance correlation coefficient, CCC) and subject-level identifiability.

ResultsMIND networks demonstrated higher edge-level HF-ULF agreement (mean CCC = 0.95) and ULF reproducibility (mean CCC = 0.98) compared to MSNs (mean CCC = 0.86-0.87 and 0.91-0.94, respectively). In contrast, MSNs yielded higher subject-level differential identifiability across field strengths due to a broader dis-tinction between within- and between-subject similarity. Reconstructing MIND networks on native high-resolution meshes (fsnative) maximised cross-scanner agreement, whereas coarser meshes and finer cortical parcellations enhanced individual network identifiability.

ConclusionsStructural similarity networks can be reliably constructed from ULF MRI data. Optimal analytic choices depend on study objectives: MIND on native meshes maximises cross-scanner stability, whereas MSNs excel at preserving individual uniqueness, paving the way for globally scalable connectomic research.

Upper cervical spinal cord morphometric charts: from population references to individualized clinical assessment
J. Wang, T. Hua, H. Wu +32
medRxiv9 Oct 2026PreprintAll rights reservedFull textMedical imaging & radiologymedRxiv
Regional brain morphometric correlates of CT-defined metabolic dysfunction-associated steatotic liver disease and hepatic fat burden.
Chongwon Pae, Minchul Kim, Jung Hoon Kim +2
BMC medical imaging4 Aug 2026Peer-reviewedCC BY-NC-NDFull textEpidemiology & public healthEurope PMCPMC
Interrater Reliability of the Modified Tuck Jump Assessment Scored by Physical Therapy Students
Makoto Saito
Cureus19 Oct 2026Peer-reviewedCC BYFull textGeneral medicineEurope PMC
An Objective Assessment for Plank Performance in Active Adults With and Without Low Back Pain: A Reliability and Comparison Study
International journal of sports physical therapy19 Oct 2026Peer-reviewedCC BY-NCFull textGeneral medicineEurope PMC
Inter‐Rater Reliability in Pre‐Anaesthetic Airway Assessment: A Protocol for a Systematic Review
Constance Thyra Giertz Christiansen, Mathilde Nellemann Jensen, Moritz Kilian German Denneborg +8
Acta anaesthesiologica Scandinavica9 Oct 2026Peer-reviewedCC BYFull textGeneral medicineEurope PMC

Cite

BibTeX RIS

BibTeX

@misc{karoui2026morphometric,
  title = {{Morphometric Brain Networks at Ultra-Low Field: A First Assessment of Feasibility and Reliability}},
  author = {H. Karoui and N. J. Bouke and J. H. Cole and R. W. Yeow and F. Dick and R. Razin and D. L. Thomas and K. H. Schroder and D. Cash and M. Bocchetta and S. C. Williams and F. Vasa},
  howpublished = {medRxiv},
  year = {2026},
  month = oct,
  doi = {10.64898/2026.10.17.26365791},
  url = {https://www.medrxiv.org/content/10.64898/2026.10.17.26365791v1},
  note = {CC BY-NC},
}

RIS

TY  - UNPB
TI  - Morphometric Brain Networks at Ultra-Low Field: A First Assessment of Feasibility and Reliability
AU  - H. Karoui
AU  - N. J. Bouke
AU  - J. H. Cole
AU  - R. W. Yeow
AU  - F. Dick
AU  - R. Razin
AU  - D. L. Thomas
AU  - K. H. Schroder
AU  - D. Cash
AU  - M. Bocchetta
AU  - S. C. Williams
AU  - F. Vasa
PY  - 2026
DA  - 2026/10/17
PB  - medRxiv
DO  - 10.64898/2026.10.17.26365791
UR  - https://www.medrxiv.org/content/10.64898/2026.10.17.26365791v1
AB  - BackgroundUltra-low field (ULF) magnetic resonance imaging (MRI) offers accessible point-of-care neuroimag-ing, yet its capability to support multivariate network analyses remains understudied.  MethodsWe systematically evaluated the feasibility and reliability of Morphometric Similarity Networks (MSN) and Morphometric Inverse Divergence (MIND) networks derived from paired high-field (HF, 3T/1.5T) and ULF (64mT) scans in a primary cohort (N=23) and an independent replication cohort (N=19). We tested the impact of super-resolution processing, cortical parcellation granularity (Desikan-Killiany vs. Destrieux), and mesh resolution (fsnative vs. fsaverage5) on edge-level cross-scanner agreement (Lins concordance correlation coefficient, CCC) and subject-level identifiability.  ResultsMIND networks demonstrated higher edge-level HF-ULF agreement (mean CCC = 0.95) and ULF reproducibility (mean CCC = 0.98) compared to MSNs (mean CCC = 0.86-0.87 and 0.91-0.94, respectively). In contrast, MSNs yielded higher subject-level differential identifiability across field strengths due to a broader dis-tinction between within- and between-subject similarity. Reconstructing MIND networks on native high-resolution meshes (fsnative) maximised cross-scanner agreement, whereas coarser meshes and finer cortical parcellations enhanced individual network identifiability.  ConclusionsStructural similarity networks can be reliably constructed from ULF MRI data. Optimal analytic choices depend on study objectives: MIND on native meshes maximises cross-scanner stability, whereas MSNs excel at preserving individual uniqueness, paving the way for globally scalable connectomic research.
ER  -

Abstract

Related

Cite