The SRI24 Multi-Channel Brain Atlas: Construction and Applications

Area of Science:

• Neuroimaging
• Human Brain Anatomy
• Medical Image Analysis

Background:

• Standard brain atlases are crucial for interpreting neuroimaging data.
• Existing atlases may have limitations in resolution, flexibility, and artifact handling.
• Advancements in magnetic resonance imaging (MRI) technology enable higher quality brain data acquisition.

Purpose of the Study:

• To develop a novel, dynamic standard atlas of the human brain.
• To improve upon existing atlases by incorporating advanced imaging techniques and data processing.
• To provide a versatile resource for various neuroimaging applications.

Main Methods:

• Generated a standard atlas using unbiased population registration of high-resolution 3T MRI data from 24 healthy subjects.
• Acquired images using multichannel-coil techniques.
• Integrated anatomical (T1-weighted, spin echo) and diffusion (fractional anisotropy, mean diffusivity, diffusion-weighted image) data, along with tissue probability maps (CSF, gray matter, white matter).

Main Results:

• The atlas is dynamic, represented by nonrigid transformations between subject images and distortion-corrected alignments.
• The atlas can be generated at arbitrary resolutions and orientations without interpolation artifacts.
• Demonstrated applications in region definition for fiber tracking (enhanced sharpness) and spatial normalization (average shape property).

Conclusions:

• The developed human brain atlas possesses unique dynamic features and superior resolution.
• Its flexibility in resolution and orientation generation makes it a valuable tool.
• The atlas will be shared with the scientific community as a reference for future brain imaging studies.