kwneuro.tractseg
================

.. py:module:: kwneuro.tractseg


Functions
---------

.. autoapisummary::

   kwneuro.tractseg.extract_tractseg


Module Contents
---------------

.. py:function:: extract_tractseg(dwi: kwneuro.dwi.Dwi, mask: kwneuro.resource.VolumeResource, response: kwneuro.resource.ResponseFunctionResource | None = None, output_type: str = 'tract_segmentation', csd_peaks: tuple[kwneuro.resource.VolumeResource, kwneuro.resource.VolumeResource] | None = None) -> kwneuro.resource.VolumeResource

   Run TractSeg on a DWI dataset to segment white matter tracts.

   :param dwi: The Diffusion Weighted Imaging (DWI) dataset.
   :param mask: A binary brain mask volume.
   :param response: The single-fiber response function. If `None`, the
                    response function is estimated automatically. Ignored when
                    ``csd_peaks`` is provided.
   :type response: Optional
   :param output_type: TractSeg can segment not only bundles, but also the end regions of bundles.
                       Moreover it can create Tract Orientation Maps (TOM).
                       'tract_segmentation' [DEFAULT]: Segmentation of bundles (72 bundles).
                       'endings_segmentation': Segmentation of bundle end regions (72 bundles).
                       'TOM': Tract Orientation Maps (20 bundles).
   :param csd_peaks: Pre-computed CSD peaks as a ``(directions, values)`` tuple
                     of VolumeResources. When provided the CSD computation step is skipped.
   :type csd_peaks: Optional

   Returns: A volume resource containing a 4D numpy array with the output of tractseg
       for tract_segmentation:     [x, y, z, nr_of_bundles]
       for endings_segmentation:   [x, y, z, 2*nr_of_bundles]
       for TOM:                    [x, y, z, 3*nr_of_bundles]