Profiling#

Once the bundles are identified, pyAFQ quantifies various properties of the tract along its length. By default, these are fractional anisotropy and mean diffusivity from diffusion tensor imaging. However, custom tissue properties may be provided (see How to add custom tissue properties from another pipeline for details) and there are a variety of other built-in tissue properties that you can find in The pyAFQ API methods.

These tract profiles are exported as a CSV. For each subject, 1-dimensional profiles are generated for each tract and each tissue property. The CSV can then be used for further statistical analysis. A useful library here is AFQ-Insight .