Researcher Sairam Geethanath’s team has built low-field MRI machines that may one day make medical imaging available to more people.

Magnetic resonance imaging has been proven as a safe, effective and often lifesaving modality of medical imaging. Unfortunately, access to this vital technology is often limited in low-income regions and countries. Sairam Geethanath, an assistant professor in the Johns Hopkins Department of Radiology and Radiological Science, is working to change that through his groundbreaking research into low-field MRI.

In 2014, Geethanath, an expert in biomedical engineering, was in India working on the country’s first 1.5 T MRI scanner — part of a national mission to advance technology. During that project, he became aware of lack of access as a critical issue affecting patients and medical professionals globally.

While MRI is a crucial component in the diagnosis and treatment of many conditions and diseases, it is sobering to note that, according to the World Health Organization’s 2017 Global Atlas of Medical Devices, a staggering 66% of the world’s population lacks access.

In the report, the WHO notes that 54% of countries globally have at least one MRI unit. American and European areas have the highest proportion of countries with specialized high-technology medical equipment, while Africa and Southeast Asia tend to have the lowest proportion of countries with access.

An evolution in MRI technology may soon change these grim statistics. Geethanath and his fellow researchers are studying low-field MRI as a potential way to bring smaller, less expensive MRI units to developing nations, low-income regions and small/unconventional physical spaces.

But what is low-field MRI? First, it is important to note that, in MRI, magnet strength is measured in units called ‘Tesla’ (T). According to Geethanath, anything less than 1 T in strength can be considered low-field MRI, with anything under .1 T being very low field.

He noted that the definition has changed as magnets continue to strengthen. Since the first MRI studies were published in 1973, magnetic fields have been pushed higher. While this leads to higher-quality images, it comes with the trade-off of larger, more expensive machines and more complex safety protocols. At Johns Hopkins, researchers work with MRI units up to 7 T in strength, though 1.5 T to 3 T is clinically standard.

Some issues are particularly suited to using low-field MRI, such as cancer, infectious diseases, musculoskeletal pathologies, neurological issues including psychiatric disorders, or pediatric conditions.

While the push for more powerful magnets has dominated for decades, the development of new technologies such as artificial intelligence and machine learning is a promising step toward better image quality using lower magnetic fields.

Currently, the challenge lies in producing clinically acceptable imaging with a low-field scanner. While low-field MRI can produce meaningful structural imaging, it is not yet clinically useful beyond anatomical imaging.

Researchers are working to change that. They can increase the signal-to-noise ratio in low-field images by using deep-learning tools to investigate the noise so it can be reduced. Artificial intelligence can also be used to learn about patterns in low-resolution images, enabling these images to be rebuilt in a process called super-resolution reconstruction.

With this promising technology, Geethanath noted, “We can start designing systems for accessibility and not just performance.”

At Johns Hopkins, Geethanath and his team are seeking approval from the Institutional Review Board to proceed with trials on patients ages 10 to 17. As low-field scanners are much smaller and have fewer safety issues than higher-field units, Geethanath’s team can design and build the entire scanner in-house.

With exciting advancements made every day, Geethanath and the researchers at Johns Hopkins are working to expand access to lifesaving medical imaging worldwide.