Scalar Image File Format

Scalar value image includes the following data types:

• brain_region: label image with integer values as voxel type

• gray_level: raw imaging data, integer or float values as voxel type

• longitude: computed depth information of the atlas, integer values as voxel type

• hemisphere: hemisphere dataset with int8 or uint8 values as voxel type (0: undefined, 1: left, 2: right)

Specification

The meta data stored in the NRRD file should follow the specification below:

Field	Value Type	Suggested Value	Comments
dimension	integer	3	dimension of the image is always 3 for volumetric data
encoding	string	gzip	use ‘gzip’ for compressing the data
endian	string	little	use ‘little’ for x86 machines
kinds	list of string	[‘domain’, ‘domain’, ‘domain’]	use ‘domain’ to indicate that the axis is image axis
sizes	list of integer	[{size_x}, {size_y}, {size_z}]	the size of the volume in x, y, z dimension
space directions	3d matrix		A 3D matrix indicating the orientation of the image data, with the value indicating the spacing of the voxel
space origin	list of float	[{origin_x}, {origin_y}, {origin_z}]	physical coordinates of the image origin
space	string	{x_orientation}-{y_orientation}-{z_orientation}	the orientation of the image data, should be consistent with the space direction
type	string	{voxel type}	the type of the values stored in the voxels (e.g. unsigned/signed int, float, etc.)

Example:

{
  u'space origin': ['-46.540000915527344', '-152.15999984741211', '-152'],
  u'space directions': [['16', '0', '0'], ['0', '16', '0'], ['0', '0', '16']],
  u'sizes': [308, 495, 464],
  u'space': 'left-posterior-superior',
  u'encoding': 'gzip',
  u'endian': 'little',
  u'kinds': ['domain', 'domain', 'domain'],
  u'type': 'unsigned short',
  u'dimension': 3
}