Researchers in Hong Kong have developed a groundbreaking ultra low-field whole-body MRI, offering a safer, quieter, and more affordable alternative to traditional high-field MRI machines. This innovative technology, using a 0.05 T magnet, requires no protective shielding and aims to improve global MRI accessibility.
For years, researchers and medical companies have explored low-field MRI systems, which have a magnetic field strength of less than 1 Tesla (T), as a feasible alternative to the loud and expensive high-field MRI machines. High-field MRI systems often require special rooms with shielding to block their powerful magnetic fields.
Most low-field scanners in development are intended for brain scans only. However, in 2022, the US Food and Drug Administration (FDA) cleared the first portable MRI system, Hyperfine’s Swoop, designed for bedside use for head and brain scans. Now, a new study published in Science describes the development of a whole-body, ultra low-field MRI by researchers from Hong Kong.
“This is a big breakthrough,” said Kevin Sheth, MD, director of the Yale Center for Brain & Mind Health. “It is one of the first demonstrations of low-field MRI imaging for the entire body.”
Innovative Technology
The device uses a 0.05 T magnet, which is sixty times weaker than the standard 3 T MRI models commonly used in hospitals today. The lead author, Ed Wu, PhD, professor of biomedical engineering at The University of Hong Kong, stated that the low magnetic field strength means no protective shielding is needed. Patients and bystanders can safely use smartphones, and the scanner is safe for patients with implanted devices like cochlear implants or pacemakers. Additionally, no hearing protection is required because the machine is so quiet.
If all goes well, the technology could be commercially available within a few years, Wu said. However, funding and FDA approval are necessary first. “A company will need to commercialize this and go through the certification process,” said Andrew Webb, PhD, professor of radiology at the Leiden University Medical Center in the Netherlands.
Improving Access to MRI
One significant hope for this technology is to bring MRI to more people worldwide. For instance, Africa has less than one MRI scanner per million residents, while the United States has about 40.
A new 3 T machine can cost around $1 million, but the low-field version is much cheaper — only about $22,000 in materials per scanner, according to Wu. The low magnetic field means less electricity is required, and the machine can be plugged into a standard wall outlet. Moreover, not needing a fully shielded room can save another $100,000 in materials, Webb said.
Its ease of use could improve accessibility in countries with limited training. “Becoming a technician for a regular MRI machine takes 2-3 years of training, a lot of which involves safety and subtle planning,” Webb noted. “These low-field systems are much simpler.”
Challenges and the Future
The prototype weighs about 1.5 tons (3000 lb), which is comparable to a mobile CT scanner. Although this might sound heavy, it is lighter than the 6-13 ton weight of a 3 T MRI machine. “Its weight can be substantially reduced with further optimization,” Wu said.
One challenge with low-field MRIs is that image quality tends to be less clear and detailed than those from high-power machines. To address this, the research team used deep learning (artificial intelligence) to enhance image quality. “Computing power and large-scale data underpin our success, tackling physics and math problems traditionally considered intractable in existing MRI methodology,” Wu explained.
Webb was impressed by the image quality shown in the study. The images “look much higher quality than you would expect from such a low-field system,” he said. However, only healthy volunteers were scanned. The true test will be using it to view subtle pathologies, which Wu and his team are currently working on. They aim to take scans to diagnose various medical conditions, similar to their brain-only version of the low-field MRI that has been used for diagnosis.