Advancing Cerebro-cerebellar Network Imaging with Ultra-high Field (7T) MRI: Progress and Future Challenges

Recent advances in MR neuroimaging: toward quantitative and AI-driven brain and spinal cord imaging.

Magnetic resonance neuroimaging is undergoing a major paradigm shift from traditional qualitative anatomical mapping toward integrated, quantitative measurement systems with biological interpretability. This review systematically synthesizes nine methodological pillars driving this transformation, encompassing advances ranging from hardware innovation to artificial intelligence algorithms. We first explore the pivotal role of deep learning in image reconstruction and acceleration, followed by detailed analyses of quantitative brain oxygen metabolism assessment, standardized spinal cord imaging frameworks, and the non-invasive monitoring of the glymphatic system using diffusion MRI. Furthermore, the review delves into tractometry, susceptibility-based myelin mapping, the clinical standardization of arterial spin labeling, and the application of radiomics in extracting high-dimensional phenotypes. Finally, the importance of open science and workflow coordination in enhancing research reproducibility is highlighted. Through the deep integration of hardware, sequences, and artificial intelligence, these technologies form a synergistic ecosystem that provides unprecedented precision tools and translational potential for both basic neuroscience research and clinical precision medicine. Across these domains, AI contributes not only to acceleration and reconstruction but also to segmentation, quality control, quantitative parameter extraction, and multiparametric pattern recognition that can support diagnostic interpretation. The quantitative emphasis of this review therefore lies in measurable outputs such as image-quality metrics, metabolic and perfusion parameters, tract-specific diffusion indices, susceptibility-based components, and radiomic features.

Ultra-Low-Field Brain MRI for Point-of-Care Applications: A Technology Review.

The recent developments in ultra-low-field brain MRI are reviewed, which enable imaging in open environments and demonstrate initial clinical applicability in point-of-care settings, and future developments are envisioned to address the current limitations of image quality and contrast in ultra-low-field brain MRI systems.

9.4-Tesla magnetic resonance imaging in focal epilepsy patients with high-resolution surface-based profiling of focal cortical dysplasias.

OBJECTIVE The detection of subtle epileptogenic lesions such as focal cortical dysplasias (FCDs) is a clinical challenge in the management of drug-resistant focal epilepsy (DRFE). Ultra-high-field (UHF) magnetic resonance imaging (MRI) offers increased signal-to-noise ratios and spatial resolution compared to 3-T MRI and may improve diagnostic yield. METHODS We recruited n = 21 DRFE patients (with 3-T MRI findings: two positive, three equivocal, 16 negative) undergoing presurgical workup and n = 20 healthy controls for 9.4-T MRI (.8 mm isotropic magnetization-prepared 2 rapid acquisition gradient echo [MP2RAGE], slabs of .375 × .375 × .8 mm T2*-weighted gradient echo) and 3-T MRI (magnetization prepared rapid acquisition gradient echo [MPRAGE], magnetization-prepared 2 rapid acquisition gradient echo [MP2RAGE], fluid-attenuated inversion recovery [FLAIR]) acquisitions. Visual review for possible epileptogenic lesions was performed by clinical experts. For histopathologically confirmed FCDs, we extracted surface-based quantitative features (cortical thickness, quantitative T1, FLAIR, T2*, and quantitative susceptibility mapping values) across cortical depths and distances from the lesion center and performed high-resolution cortical profiling of 9.4-T T2* values. RESULTS In two patients with histopathologically confirmed FCD IIb, lesions were visible with distinct qualitative and quantitative features at both field strengths. One of these type IIb FCDs showed a focal cortical T2* reduction at 9.4 T that could be quantified via automated cortical profiling, consistent with the previously described "black line sign." No new epileptogenic lesions were identified at 9.4 T in 3-T MRI-negative patients, who also had no histological evidence of such lesions. SIGNIFICANCE 9.4-Tesla MRI findings in epileptogenic lesions underlying DRFE are consistent with those on 3-T MRI. UHF T2*-weighted sequences may be useful to detect the black line sign and thereby refine surgical or ablation targeting for some FCDs. Assessment of the diagnostic yield of 9.4-T MRI was limited by the lack of 3-T MRI-negative but histopathologically confirmed cases and by the unavailability of parallel transmit and FLAIR at 9.4 T. Further optimization of UHF protocols and analysis methods on larger cohorts may enhance clinically applicable diagnostic benefits.

Combined Metabolic and Microstructural Tractometry of the Superior Longitudinal Fasciculus in Healthy Brains: A Proof-of-Concept Study

ALFIE: Anatomy-aware enhancement of Low FIEld 64mT T2-weighted neonatal brain MRI for structural analysis

Purpose: To develop and evaluate an anatomy-aware deep learning framework for enhancement of neonatal 64mT T2-weighted MRI that improves anatomical visibility while preserving native ultra-low-field contrast and enabling quantitative structural analysis. Methods: A multitask network, jointly performing image enhancement and tissue segmentation, was trained on 75 and evaluated on 20 paired neonatal 64mT/3T MRI datasets spanning a broad range of gestational ages and pathologies. To preserve native 64mT contrast, 3T images were locally harmonized before training. The framework also generated quality-control maps and regional volumetric measurements. Volumetric agreement was further assessed in 40 paired term-born control datasets. Results: Enhanced 64mT images showed improved image quality metrics and better delineation of cortical, deep gray matter, ventricular, white matter, and posterior fossa structures while maintaining native contrast characteristics. Tissue segmentations demonstrated good agreement with reference 3T labels. Volumetric measurements showed excellent correspondence with 3T across major tissue compartments, with only small systematic regional biases. Conclusions: Anatomy-aware enhancement enables automated tissue segmentation and volumetric analysis directly from neonatal 64mT MRI while preserving native image contrast. These findings support the feasibility of quantitative neonatal neuroimaging at ultra-low field.

Hydrocephalic Brain Volume Estimation from Low-Field MRI: Topologically-Enriched Cross-Modal Enhancement and Segmentation

Objective: Accurate volumetric analysis of the brain and cerebrospinal fluid (CSF) is essential for monitoring hydrocephalus, a significant pediatric neurological condition. While computed tomography (CT) provides high-quality volumetric assessment, its associated ionizing radiation poses risks, especially for children. Low-field magnetic resonance imaging (LF-MRI) offers a safer and more accessible alternative, particularly in resource-constrained settings. However, its lower resolution and increased susceptibility to structural distortions make accurate segmentation challenging. This study aims to demonstrate that reliable volumetric measurements can be obtained from LF-MRI that are comparable to CT, enabling safer and more frequent monitoring of infants with hydrocephalus. Approach: We propose EnSegNet-Cross, a cross-modality, enhancement-aware segmentation network for brain volume analysis using LF-MRI. The framework leverages high-fidelity CT data during training but requires only LF-MRI during inference. At the core of the framework is a novel cross-modal topological penalty designed to minimize discrepancies between predicted LF-MRI and CT structures. A central contribution is the integration of a three-dimensional topological loss based on persistent homology, which penalizes topological discrepancies in CSF regions, specifically CSF holes formed by enclosed brain parenchyma, between CT and LF-MRI segmentations. Incorporating these structural priors facilitates generalization across heterogeneous clinical cases while eliminating the need for CT data during inference, resulting in more anatomically coherent and topologically faithful segmentations. Main Results: On a curated cohort of infants with hydrocephalus who had paired LF-MRI and CT scans, including cases with infectious and non-infectious causes, EnSegNet-Cross consistently outperformed state-of-the-art machine learning alternatives. It achieved the highest Dice score of 0.8532 plus/minus 0.03 and Volume Score of 0.9318 plus/minus 0.03. The method also demonstrated robust performance in challenging cases with confounding factors, achieving a Dice score of 0.8340 plus/minus 0.03 and a Volume Score of 0.9111 plus/minus 0.05. By leveraging CT-derived topological priors, EnSegNet-Cross successfully handled anatomically complex scenarios in which conventional models failed. Significance: EnSegNet-Cross provides a reliable and interpretable solution for brain and CSF segmentation, particularly in complex cases of hydrocephalus. This study demonstrates that high-fidelity volumetric estimates can be achieved using only LF-MRI, facilitating frequent, radiation-free monitoring. By bridging the fidelity gap between low-quality LF-MRI and high-resolution CT through clinically grounded enhancement and topological supervision, EnSegNet-Cross offers a robust clinical tool for brain volumetric analysis in infants with hydrocephalus using LF-MRI.