TY  - CHART
AU  - Schiffer, Christian
AU  - Kedo, O.
AU  - Amunts, Katrin
AU  - Dickscheid, Timo
TI  - Ultrahigh resolution 3D cytoarchitectonic map of the human amygdala created by a Deep-Learning assisted workflow (v1)
PB  - EBRAINS
M1  - FZJ-2022-04994
PY  - 2022
AB  - This dataset contains automatically created detailed map of 13 cytoarchitectonic subdivisions of the amygdala and 6 fiber bundles in the BigBrain dataset. The mappings were created using Deep Convolutional Neural Networks based on Schiffer et al 2021, which were trained on delineations on at least every 15th section created based on Kedo et al 2018. Mappings are available on every section. Their quality was observed by a trained neuroscientist to exclude sections with low quality results from further processing. Automatic mappings were transformed to the 3D reconstructed BigBrain space using transformations used in Amunts et al 2013, which were provided by Claude Lepage (McGill). Mappings on individual sections were used to assemble 3D volumes of all areas. Low quality results were replaced by interpolation between nearest neighboring sections. The volumes were then smoothed using a 3D median filter and largest connected components were identified to remove false positive results of the classification algorithm. The dataset consists of a HDF5 file containing the volume in RAS dimension ordering (20-micron isotropic resolution, dataset “volume”) and an affine transformation matrix (dataset “affine”). An additional dataset “interpolation_info” contains an integer vector with an integer value for each section which indicates if a section was replaced by interpolation due to low quality results (value 2) or not (value 1). Due to the large size of the volume, it is recommended to view the data online using the provided viewer link.
KW  - Neuroscience (Other)
LB  - PUB:(DE-HGF)32
DO  - DOI:10.25493/TKTP-7NR
UR  - https://juser.fz-juelich.de/record/911742
ER  -