mattcieslak / dsi2

Local termination pattern analysis in python

DSI2

DSI2 provides all the features necessary to perform local termination pattern analysis. Your data needs to be reconstructed using other tools, but once tractography in the TrackVis format is ready, DSI2 enables rapid prototyping of streamline clustering algorithms and group streamline analysis.

Documentation: http://dsi2.readthedocs.org

Credits

• Matt Cieslak (mcieslak)
• Matt Jaswa (mjaswa)
• Jessica Huang (justtjess)

dsi2's Issues

Remove doc/build from repository

Develop an optimal document structure for MongoDB backend

Should there be a collection for each parcellation scheme?

Should each subject have a single document containing the voxel mapping?

Should authentication be built into the DB? Or do we assume that users will be accessing data locally?

What should we do about the actual streamlines? Use GridFS? What about PyTables?

Document how to make a new Aggregator

Create a setup.py that works on multiple platforms

We should have something that works with pip or easy_install to install dsi2 as a library

Configure TrackDataSource to work with MongoDB

Currently TrackDataSource only works with TrackDatasets that have been load into memory. This is a big limitation as the database grows.

A preliminary version of a MongoDB backend is in dsi2.database.tables and dsi2.database.parallel.

Document how to use the GUI (with screencaps)

DSI2

dsi2.volumes.scalar_volume.ScalarVolume has a C++ error when using wx backend

"Underlying c++ object has been deleted"

Convert print statements to logging

The gui loops all have print statements that should be converted to logging commands.

Make screen shot menu obey the magnification and DPI settings

These settings are currently ignored