dHCP Brain Atlas 2017

Neonatal brain atlas constructed from 275 term-born subjects of the dHCP cohort in 2017 using the "groupwise" approach described in Schuh (2017) (PhD thesis, ch. 6.3). This release version of BAtCh contains the global normalization input files and configuration files for the mirtk construct-atlas and mirtk evaluate-atlas scripts which are part of MIRTK. The affine transformations used for global normalization of the brain scans were computed using the "pairwise" HTCondor workflow (cf. dhcp-v2.0).

The used method is described in:

• A Schuh, A Makropoulos, E C Robinson, L Cordero-Grande, E Hughes, J Hutter, A Price, M Murgasova, R P A G Teixeira, N Tusa, J Steinweg, S Victor, M A Rutherford, J V Hajnal, D A Edwards, D Rueckert, Unbiased construction of a temporally consistent morphological atlas of neonatal brain development, bioRxiv (2018).
• A Schuh, Computational models of the morphology of the developing neonatal human brain, PhD thesis, ch. 6.3 (2017).

The constructed atlas is available under DOI 10.12751/g-node.d2b353.

dHCP Brain Atlas 2017 (pairwise)

Neonatal brain atlas constructed from 275 term-born subjects of the dHCP cohort in 2017 using the "pairwise" approach with SVFFD transformation model, an inverse-consistent symmetric registration energy, and Log-Euclidean mean deformations. This version of the atlas construction workflow has been used to obtain the results for Schuh (2017) (PhD thesis, ch. 6.2).

The initial workflow steps implement the "Unbiased global normalization" as described in chapter 6.1 of Schuh (2017). The thus obtained affine output transformations were also used for the non-rigid atlas construction using the more recent iterative "groupwise" approach (cf. dhcp-v3.0).

The used method is described in:

• A Schuh, M Murgasova, A Makropoulos, C Ledig, SJ Counsell, JV Hajnal, P Aljabar, D Rueckert, Construction of a 4D Brain Atlas and Growth Model using Diffeomorphic Registration, 3rd MICCAI Workshop on Spatiotemporal Image Analysis for Longitudinal and Time-Series Image Data (STIA), LNCS 8682, pp. 27-37, Boston, MA (2014).
• A Schuh, Computational models of the morphology of the developing neonatal human brain, PhD thesis, ch. 6.2 (2017).

Preterm Brain Atlas 2014

First original brain atlas construction workflow following the pairwise approach similar to Serag et al. (2012), but using SVFFD transformation model with symmetric energy function and Log-Euclidean mean deformations. This method has been developed pre-dHCP using earlier preterm brain scans (cf. Git tag preterm). This workflow can also be used to replicate the method Serag et al. (2012) by using the appropriate configuration file.

The used method is described in:

• A Schuh, M Murgasova, A Makropoulos, C Ledig, SJ Counsell, JV Hajnal, P Aljabar, D Rueckert, Construction of a 4D Brain Atlas and Growth Model using Diffeomorphic Registration, 3rd MICCAI Workshop on Spatiotemporal Image Analysis for Longitudinal and Time-Series Image Data (STIA), LNCS 8682, pp. 27-37, Boston, MA (2014).