release 0.7.1

Added compatibility with both nilearn 0.8 and 0.9+

release 0.7

New feature: it is now possible to filter out small clusters from the dypac analysis. By default clusters that are smaller than 50% of the cluster size of a uniform solution (i.e. all culsters have same size) get excluded. This removes some noise as well as estimates of dwell times.

release 0.6

Addition of two masker classes: LabelsMasker and MapsMasker, and accompanying tutorial notebooks.
These classes can be used to generate dypac-style maskers for any type of hard or soft parcellation.

What's Changed

• Update nilearn version to 0.9.0 by @htwangtw in #85
• Iss83 masker by @pbellec in #86

New Contributors

• @htwangtw made their first contribution in #85

Full Changelog: v0.5.3...v0.6

release 0.5.3

Bug fix release: the confounds were not properly handled by MultiNiftiMasker, and not regressed out of the data prior to estimating the parcels.

release 0.5.2

Grey matter segmentation is now used to constrain the brain mask, instead of re-weighting signals.
This is accessible through the grey_matter and threshold_grey_matter arguments, and can be skipped by specifying grey_matter=None.

release 0.5.1

Minor update:

• consensus clustering using k_means now uses the sample_weight argument, set to the dwell times of the parcellations being aggregated.

release 0.5

This release includes:

• major bug fixes.
• main class in dypac is now named Dypac.
• working batch system to limit memory / computing load.
• weighting of grey matter for improved detection of RSNs. Check parameters grey_matter and std_grey_matter.
• embedding tools to transform to and from the parcellation space. Check new attributes transform, inverse_transform, compress and score.
• new demo dypac_demo.ipynb
• many bug fixes.

Contributors:
Hanad Shamarke, Désirée Lussier have contributed bug reports and optimization.
François Paugam contributed the initial version of the embedding tools.
Pierre Bellec implemented bug fixes, batching, grey matter weights and embeddings.

refactoring

There are now two modules:

• bascpp features the core "barebone", generic methods.
• dypac has the object class for fMRI data

The method has also been changed: now stab_maps are extracted and trimmed within batches, and then matched and averaged across batches by a separate consensus clustering. Hopefully this approach will scale very well.

Important change in the API: n_states now controls the total number of states, which before it was the number of states per cluster.

initial release

The main method dypac is functional, and the notebook dypac_fmri_compression.ipynb gives a demo of the method.