Related Experiment Videos
Quantitative 1H spectroscopic imaging of human brain at 4.1 T using image segmentation
H P Hetherington1, J W Pan, G F Mason
1Department of Medicine, University of Alabama at Birmingham, USA.
Magnetic Resonance in Medicine
|July 1, 1996
Summary
This study quantifies metabolic differences in brain tissue using 1H spectroscopic imaging. The findings help accurately interpret brain metabolite levels, aiding in the diagnosis of neurological conditions like multiple sclerosis.
Area of Science:
- Neuroimaging
- Biochemistry
- Medical Physics
Background:
- Metabolic differences in N-acetylaspartate (NAA), creatinine (CR), and choline (CH) between cerebral gray and white matter complicate 1H spectroscopic image interpretation.
- Accurate interpretation requires knowledge of the gray- and white-matter composition within each voxel.
Purpose of the Study:
- To develop and validate a method for determining voxel tissue composition at 4.1 Tesla.
- To establish reliable reference values for brain metabolites (CR/NAA, CH/NAA) accounting for tissue variations.
- To identify metabolic abnormalities in neurological disease using this refined spectroscopic approach.
Main Methods:
- Implemented a T1-based image segmentation scheme at 4.1 Tesla to determine voxel tissue composition.
- Utilized the point-spread function of spectroscopic imaging and segmented anatomical images to define tissue composition.
- Performed linear regression analysis on 984 voxels from 10 subjects, using white-matter CR as an internal standard, to determine pure gray and white matter metabolite values.
- Established means and confidence intervals for CR/NAA and CH/NAA ratios for voxels with arbitrary tissue composition.
Main Results:
- Successfully determined the tissue composition of each voxel by integrating spectroscopic imaging point-spread function with segmented anatomical data.
- Quantified pure gray- and white-matter values for CR/NAA and CH/NAA ratios, along with individual metabolite concentrations.
- Established a method to derive reliable CR/NAA and CH/NAA values for voxels of mixed tissue composition.
- Identified the extent and location of metabolic abnormalities (P < 0.05) in a patient with multiple sclerosis using the developed method.
Conclusions:
- The T1-based segmentation and spectroscopic analysis method accurately accounts for gray and white matter metabolic variations.
- This approach enables more precise interpretation of 1H spectroscopic images, improving the detection of metabolic abnormalities in neurological disorders.
- The established reference values and methodology provide a robust tool for clinical neuroimaging research and diagnostics.