Advancing MR Fingerprinting for Ultra-High-Field Scanners to Map the Brain
Since its invention in the early 1970s, MRI has become a mainstay of medical imaging, with approximately 40 million MRI scans performed each year in the United States. Characterized by its exceptional ability to visualize soft tissues in detail, ability to see structures hidden behind bone, and lack of ionizing radiation, MRI has become the preferred imaging modality for seeing structures in the brain.
As the MR technology matured, new MR techniques have also been developed.
“Fingerprinting” Tissues with MRI
In recent years, MR Fingerprinting has emerged as an advanced MR technique that allows for quantitative imaging by measuring multiple tissue properties at the same time, during a single scan. MR Fingerprinting deliberately varies the scanner settings and imaging parameters. As these settings change, each type of tissue responds in its own way over time, creating a unique signal pattern, like a fingerprint.
In a new study from Kawin Setsompop, PhD, Professor of Radiology (RSL), the research team tailored the MR Fingerprinting technique for next-generation 7T MRI scanners at Berkeley, using a similar system that was installed at the Lucas Center last year. The high magnetic field and superior gradient performance of the new scanner enables higher resolution but also brings its own set of challenges, including magnetic field inhomogeneities, radiofrequency distortions, and motion sensitivity.
Dr. Setsompop’s team developed an MR Fingerprinting technique for ultra-high-field MR systems that addresses all of these concerns while keeping scan times short.
As published in a recent paper, the research team used the MR Fingerprinting technique to produce whole-brain quantitative maps with high enough resolution to see tiny brain structures that would typically be difficult to visualize in living people.
Kawin Setsompop, PhD
Conventional MRI scanners produce images with voxels – or three-dimensional “pixels” – measuring roughly 1 mm on each side. With the new MR Fingerprinting technique, Dr. Setsompop’s team were able to reduce each edge of the voxel down by half, to 560 micrometers, and image at ~ 6-fold smaller voxel volume. By extending the scan time, they could improve this even further, achieving 360 micrometer isotropic resolution.
A New Window into the Brain
Figure caption: Whole‐brain T1 maps (left two columns) and T2 maps (right two columns) at 360‐μm isotropic resolution. The red box highlights the hippocampus. The red arrow indicates the exceptional image quality and structural detail achieved in the ultra‐high‐resolution quantitative maps. Black arrows, from top to bottom, point to the optic radiations, cerebellar vermis, and pontes grisei. The yellow arrows point to deep‐brain nucleis.
Contrast‐weighted T1 MPRAGE, T2W, DIR, and FLAIR images synthesized from the quantitative maps using the proposed method.
Each voxel in an image not only comprises the larger visual image but also carries precise measurements of tissue properties. These properties translate into information that researchers can use to detect disease earlier, more reliably monitor disease progression, and develop new imaging biomarkers.
Together, the tissue properties conveyed by the quantitative maps and the corresponding anatomical images provide a comprehensive look into the brain. In fact, the study revealed that the MR Fingerprinting technique for multi-parametric quantitative imaging can help optimize conventional contrast-weighted MR sequences to maximize contrast for specific brain structures. This would allow clinicians to gain a more accurate look at brain structures such as the substantia nigra and subthalamic nucleus – two small structures deep in the brain that are critically important to movement and neurodegenerative disease and are notoriously difficult to image.
Shaping MR Fingerprinting for Ultra–High-Field MRI
Beyond the technical achievements, the study also highlighted an important consideration: despite ultra-high-field MRI scanners offering advantages like higher signal, the associated challenges cannot be ignored and require specific modifications to existing MR Fingerprinting approaches.
Dr. Setsompop’s team designed a multitude of corrections and mitigation approaches as part of a proposed MRF pipeline with seamless usability. The new MR Fingerprinting approach delivers fast, high-resolution quantitative brain imaging that makes detailed tissue measurements practical for both clinical and neuroscience research. By pushing the resolution limits of multi-parametric quantitative MRI, this work opens new possibilities for studying subtle, fine-scale changes in the living human brain.