A research team led by Professors Namkyoo Park, Sunkyu Yu, and Jongho Lee (B) has proposed a new technology that addresses the longstanding challenges of image inhomogeneity and tissue heating in ultra-high-field magnetic resonance imaging (MRI) operating at 7 tesla or higher. The researchers developed a metasurface design algorithm based on electromagnetic scattering theory, successfully achieving a uniform magnetic field distribution across the entire brain.

MRI is a widely used non-invasive medical imaging technology that enables observation of internal anatomical structures and physiological changes. Higher static magnetic field strength (B0) improves the signal-to-noise ratio (SNR) and spatial resolution, allowing more precise diagnosis of complex neurological diseases such as Alzheimer’s disease and Parkinson’s disease. Although the recent development of ultra-high-field MRI has significantly enhanced brain imaging precision, it has also introduced new physical challenges.

The primary issue arises from the wavelength of the RF magnetic field (B1+). As magnetic field strength increases, the RF frequency rises and the wavelength inside human tissue becomes shorter. When this wavelength becomes comparable to the size of the human head, complex electromagnetic interference occurs. As a result, the B1⁺ field—responsible for generating MRI signals—tends to concentrate at the center of the brain while weakening at the periphery, leading to non-uniform images. In addition, localized concentration of electromagnetic energy can increase the specific absorption rate (SAR), causing unwanted tissue heating.

Conventional solutions have relied on two main approaches. One involves passive methods, such as using high-permittivity pads or metallic structures to locally adjust RF fields. The other is parallel transmission (pTx), which employs multiple RF transmit coils to control phase and amplitude. However, passive methods struggle to achieve uniformity across the entire brain, while pTx systems require complex hardware and control mechanisms and pose additional safety management challenges.

To overcome these limitations, the research team proposed a new approach using a phase-controlled metasurface. A metasurface consists of arrays of artificial subwavelength structures that can precisely manipulate the phase and wavefront of electromagnetic waves. By applying an optimization algorithm based on scattering theory, the researchers designed a metasurface that rearranges RF waves with a minimal number of elements, enabling the formation of a uniform B1+ field throughout the brain.

One notable advantage of this technology is that it can be implemented without modifying the hardware of existing MRI systems. The team validated its performance using commercial electromagnetic simulation software with multiple human brain models, including male, female, and pediatric anatomies.

The results showed that applying the metasurface improved B1+ field uniformity by approximately twofold. The coefficient of variation (CV), a metric for field uniformity, decreased from 0.32 to 0.16 on average. At the same time, the SAR value decreased by about 23%, indicating improved safety.

The researchers explained that the study applies the concept of “constant intensity waves,” previously explored in optics, to the problem of electromagnetic field control in MRI. This approach is expected to enable more uniform and safer brain imaging in ultra-high-field MRI, ultimately improving the diagnostic accuracy for neurological disorders.

Volumetric B1+ field homogenization in 7 Tesla brain MRI using metasurface scattering.

Gyoungsub Yoon, Sunkyu Yu*, Jongho Lee & Namkyoo Park*

ACS Photonics, https://doi.org/10.1021/acsphotonics.5c02781, February 2026

Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=57568

Translated by: Changhoon Kang, English Editor of the Department of Electrical and Computer Engineering, changhoon27@snu.ac.kr