Ultra-Low-Field MRI for Dementia Assessment: A Scoping Review

Background: Structural neuroimaging is a cornerstone of dementia assessment, yet timely access remains a major global challenge, often constrained by hospital-based pathways and long waiting times. Ultra-low-field (ULF) MRI is an emerging imaging technology that combines low magnetic field strength with advanced image reconstruction and processing methods to generate clinically interpretable brain images. ULF-MRI scanners are small, portable, less enclosed, operate from a standard power socket, and do not require specialist cooling or shielding systems. These features create new opportunities to deploy neuroimaging in more accessible settings. We present findings from a scoping review synthesising current evidence on the use of ULF-MRI in contexts relevant to dementia assessment.

Methods: The review followed the PRISMA Extension for Scoping Reviews (PRISMA-ScR) and was registered on the Open Science Framework. Five electronic databases were searched using predefined terms relating to ULF-MRI. Eligible studies focused on non-acute adult neuroimaging and included comparisons between ULF-MRI and high-field MRI. Studies addressing acute presentations, including traumatic brain injury or cerebrovascular events, were excluded. Study characteristics and outcomes were extracted using a standardised framework. Study selection and data extraction were undertaken independently by two reviewers.

Results: 2,015 unduplicated records were identified, from which 24 full-text articles were retrieved and 14 studies met inclusion criteria. Included studies involved healthy participants and individuals with neurological conditions, including Alzheimer’s disease, multiple sclerosis, and epilepsy. All studies used the Hyperfine Swoop 0.064T scanner and examined brain morphometry and/or white matter hyperintensities. Two studies relied exclusively on visual assessment, while the remainder used quantitative approaches, most commonly super-resolution reconstruction followed by automated tissue segmentation. Quantitative analyses demonstrated strong correlations between ultra-low-field MRI and conventional MRI for both global and regional brain volume measures, including white matter metrics, although performance varied by brain region and analysis pipeline.

Conclusions: Evidence from this review suggests that ultra-low-field MRI can support both expert visual assessment and quantitative measurement of structural brain changes relevant to dementia, including global and regional brain volume and white matter pathology. ULF-MRI shows promise as an accessible neuroimaging approach with potential to reduce diagnostic delays and inequalities in dementia assessment pathways.