v0.3.5: brain_qa

Changes:

• Fix: Generate documentation for latest modules (braindescriptors, brainlocate, etc.)
• Drop official support for Python 2.7
• Add new command line tools, including one to perform file-level QA on FreeSurfer data
• expose more functions in general API
• Add more documentation to workflow document

v0.3.4: Brain descriptors

• Added functions to get custom stats for a measure and atlas region
• Added new convenience functions for loading subject data in native and standard space
• Added BrainDescriptors class to quickly gather all brain descriptors for a complete dataset

v0.3.3: Find voxel segmentation classes, apply transforms, parse m3z files

• Added support for loading affine transformation matrices, e.g., files or mri_info output that follows the FreeSurfer register.dat format.
• New group-level functions for loading stats data
• Support for parsing talairach.xfm files
• Add brainlocator class that allows to conveniently find the vertex in a surface file that is closest to some RAS coordinate.
• Add BrainVoxLocator class that allows to conveniently find the segmentation label of a voxel. Also supports finding the label of the closest voxel with non-empty (not 'Unknown') label and gives the distance to it.
• Add support for loading m3z files, like mri/transforms/talairach.m3z

v0.3.2: Export brain meshes!

Version 0.3.2

• Added support for exporting meshes to formats used by standard 3D modeling software like Blender in formats OBJ (Wavefront object) and Stanford PLY.
• Improve some doc strings
• Add more unit tests

v0.3.1: Support for annotations and labels

Version 0.3.1

• Added support for loading FreeSurfer annotation files. This allows one to load cortical parcellation information from an Atlas, e.g., from the Desikan atlas from the '?h.aparc.annot' files. An annotation consists of several groups of vertices, each of which is assigned a label and a color.
• Added support for loading FreeSurfer label files. This allows one to load sets of vertices assigned to a group, e.g., the list of all vertices which belong to the cortex from the files '?h.cortex.label'.
• Add support for masking data using an arbitrary mask, i.e., set the data values to NaN at a list of indices. Typically this is used to mask out the medial wall in brain morphometry data. You can use any data, e.g., a loaded label, to build the mask.

Brainload 0.3.0: Python 3 Support

• Added support for Python 3. We now test for Python 2.7, Python 3.5 and Python 3.6 on Travis CI.
• Renamed all occurrences of the word morphology to morphometry, including in meta data entries and function names. This breaks compatibility with v0.2.0.
• Extend documentation: add workflow examples. The documentation is no longer only API documentation.
• Add recipe for anaconda build.

Brainload 0.2.0: Load groups of brains!

Brainload v0.2.0 is here and brings support for loading group data.

If you follow the FreeSurfer convention to have all your subject data in a single directory defined by the environment variable SUBJECTS_DIR, the new high-level functions make it very easy to load data for group comparison. See the entry for brainload.group() function in the API documentation for examples.

The binary here is identical to the one available on PyPI, and you should be able to install it using pip:

pip install --user brainload

Or use a virtual environment.