Portable MRI Shows Promise for Expanding Brain Imaging for Alzheimer’s Disease

By MedImaging International staff writers
Posted on 13 Dec 2024

Image: Application of LF-SynthSR v2 and WMH-SynthSeg to a cohort of MCI/AD subjects imaged in the clinic at LF (Photo courtesy of Nature Communications; doi.org/10.7910/DVN/9PANMC)

By 2050, an estimated 139 million people globally are expected to have Alzheimer's disease (AD). Magnetic resonance imaging (MRI) plays a key role in detecting changes in brain structure that precede cognitive decline and disease progression; however, its high cost limits its accessibility. Now, a new study has shown that a simplified, low magnetic field (LF) MRI machine, enhanced with machine learning tools, provides comparable results to traditional MRI in assessing brain characteristics associated with AD. The research, published in Nature Communications, underscores the potential of LF-MRI to evaluate individuals with cognitive symptoms.

The study team, consisting of clinical researchers, MRI physicists, health system delivery experts, and AI specialists from Mass General Brigham (Somerville, MA, USA), has been investigating LF-MRI as an alternative to high-field (HF) MRI for several years. HF-MRI uses powerful magnetic fields to generate high-resolution cross-sectional images of the body but requires expensive equipment and designated imaging facilities, often unavailable in low-resource settings both in the U.S. and worldwide. In contrast, LF-MRI machines operate with magnetic fields 50 times weaker than those used in conventional MRI. This results in smaller, portable scanners that only require a single electrical outlet to operate, but at the expense of lower image quality.

To enhance LF-MRI image quality and make it more practical for clinical use, the researchers integrated artificial intelligence (AI) tools. They generated artificial datasets by matching HF- and LF-MRI scans from both healthy individuals and patients with neurological conditions. These datasets were used to train an AI algorithm to identify AD-related features on LF-MRI, such as the size of the hippocampus (the brain's memory center) and regions of white matter hyperintensity (WMH), which may indicate neurodegeneration or vascular issues.

When tested on 54 patients with mild cognitive impairment or AD-related dementia, the AI-enhanced LF-MRI scans showed results closely matching those from traditional HF-MRI in terms of hippocampus size and white matter volume. While the new method requires regulatory approval and the development of new clinical protocols, it has the potential to expand access to neuroimaging in areas with limited MRI resources. In addition to improving AD diagnosis, LF-MRI could streamline care for AD patients needing MRI monitoring during treatment with new AD therapies. Its portability also makes it suitable for use in emergency rooms, community health centers, and ambulatory units, particularly for patients at risk of or recovering from stroke.

“Access to traditional MRI is not evenly distributed and not available to everyone,” said senior author W. Taylor Kimberly, MD, PhD, chief of the Division of Neurocritical Care in the Department of Neurology at MGH. “We envision a future where a person with cognitive complaints visiting a primary care physician, geriatrician or neurologist can get a brain scan, a blood test and a cognitive test, all in a single visit. Low-cost, easier to deploy technology can help provide information to clinicians, right at the bedside.”