Construction of a consistent high-definition spatio-temporal atlas of the developing brain using adaptive kernel...

Area of Science:

• Neuroimaging
• Developmental Neuroscience
• Medical Image Analysis

Background:

• The early developmental period witnesses significant brain changes in size, shape, and appearance.
• A comprehensive spatio-temporal atlas is crucial for understanding dynamic brain development in neonates.

Purpose of the Study:

• To develop a 4D (spatio-temporal) atlas of the developing brain for premature neonates (28-44 weeks post-menstrual age).
• To enhance anatomical definition and population representation compared to existing atlases.

Main Methods:

• Utilized T1 and T2 weighted MR images from 204 premature neonates.
• Employed non-rigid registration between all image pairs to minimize atlas bias.
• Applied kernel regression with a time-varying kernel width for age-dependent templates.

Main Results:

• The resulting 4D atlas exhibits superior anatomical definition compared to atlases created with affine or standard non-rigid registration.
• The atlas accurately reflects both global and local brain changes, serving as a good population representative.
• Demonstrated consistent detail across all time-points due to the novel time-varying kernel width.

Conclusions:

• The developed 4D brain atlas offers enhanced anatomical detail for improved registration with individual subjects.
• This atlas provides a valuable tool for studying brain development in premature infants.
• The atlas is publicly accessible, facilitating further research in developmental neuroscience.