More information: UNC 4D longitudinal infant surface atlases of cortical structures from neonates to 6 years of age contain 11 time points, including 1 month, 3 months, 6 months, 9 months, 12 months, 18 months, 24 months, 36 months, 48 months, 60 months and 72 months, thus densely covering and well characterizing the critical stages of the dynamic early brain development.

Will the version 2.0 include deep learning for brain segmentation or skull stripping?

Originally posted by meiling wang:
Yes, of course since deep learning is current state-of-the-art. :)

What do you think of deep learning ? In addtion to deep learning, what other ways to segment the brains?

Will the infant brain atlases be publicly avaiable, espcially for the 6 months of age?

Originally posted by jieyu zhang: Thanks for very much for your questions.

There are many methods for segmentation, e.g., multi-atlas based label fusion, LINKS (https://www.ncbi.nlm.nih.gov/pubmed/2554...).
Recently, deep learning based methods, e.g., ResNet, Densenet, SegNet, have achieved state-of-the-arts performance in many applications.
Therefore, we will employ these deep learning based methods in the new version.

Originally posted by lgn_njust: Yes, we will provide atlases for 2 weeks, 3, 6, 9, 12, 18, and 24 months.