High-Resolution Brain Metabolic Imaging at Ultrahigh Field Using Extended Spatiospectral Encoding and Subspace...

Area of Science:

• Neuroimaging
• Metabolic Imaging
• Magnetic Resonance Spectroscopy

Background:

• Accurate mapping of human brain metabolite distributions is crucial for understanding neurological disorders.
• Ultrahigh field (7T) magnetic resonance imaging (MRI) offers enhanced sensitivity and resolution for in vivo studies.
• Existing metabolic imaging techniques face challenges in achieving high spatial resolution and quality at ultrahigh fields.

Purpose of the Study:

• To develop and validate a high-resolution magnetic resonance metabolic imaging method for human brain metabolite mapping at 7T.
• To achieve isotropic nominal resolutions of 3.0 mm and 1.8 mm within clinically feasible scan times.
• To overcome challenges such as spectral ghosting, low signal-to-noise ratio, and spectral aliasing.

Main Methods:

• Implemented a free-induction-decay (FID) based MR spectroscopic imaging (MRSI) sequence utilizing fast echo-planar spectroscopic imaging (EPSI) trajectories.
• Acquired 3D MRSI data at 3.0 mm and 1.8 mm isotropic nominal resolutions in 4.8 and 14.2 minutes, respectively.
• Developed model-based processing methods incorporating subspace learning, spectral modeling, and generalized series modeling.

Main Results:

• Successfully generated high-resolution, high-quality metabolite maps of the human brain at 7T.
• Validated the proposed method using phantom and in vivo scans.
• Demonstrated the capability to capture detailed brain metabolite distributions with the developed technique.

Conclusions:

• The developed MRSI acquisition and model-based processing methods enable high-resolution brain metabolic imaging at ultrahigh field (7T).
• This method provides a feasible and powerful tool for investigating brain metabolism.
• The technique holds promise for various brain imaging applications, aiding in the study of neurological conditions.