{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "208da207",
   "metadata": {},
   "source": [
    "# Using RecoBundles for bundle recognition\n",
    "\n",
    "For bundle recognition, pyAFQ defaults to use the waypoint ROI approach\n",
    "described in [^Yeatman2012]. However, as an alternative approach, pyAFQ also\n",
    "supports using the RecoBundles algorithm [^Garyfallidis2017], which uses an\n",
    "atlas of bundles in streamlines. This example shows how to\n",
    "use RecoBundles for bundle recognition.\n",
    "\n",
    "The code closely resembles the code used in the GroupAFQ tutorial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e343b254",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os.path as op\n",
    "import AFQ.data.fetch as afd\n",
    "from AFQ.api.group import GroupAFQ\n",
    "import AFQ.api.bundle_dict as abd\n",
    "\n",
    "afd.organize_stanford_data(clear_previous_afq=\"track\")\n",
    "\n",
    "tracking_params = dict(n_seeds=25000,\n",
    "                       random_seeds=True,\n",
    "                       rng_seed=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "696ede38",
   "metadata": {},
   "source": [
    "## Defining the segmentation params\n",
    "We also refer to bundle recognition as the \"segmentation\" of the tractogram.\n",
    "Parameters of this process are set through a dictionary input to the\n",
    "`segmentation_params` argument of the GroupAFQ object. In this case, we\n",
    "use `abd.reco_bd(16)`, which tells pyAFQ to use the RecoBundles\n",
    "algorithm for bundle recognition. This uses 16 bundles, there is also\n",
    "an atlas `abd.reco_bd(80)` which uses 80 bundles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0a8437a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "myafq = GroupAFQ(\n",
    "    output_dir=op.join(afd.afq_home, 'stanford_hardi', 'derivatives',\n",
    "                       'recobundles'),\n",
    "    bids_path=op.join(afd.afq_home, 'stanford_hardi'),\n",
    "    # Set the algorithm to use RecoBundles for bundle recognition:\n",
    "    bundle_info=abd.reco_bd(16),\n",
    "    dwi_preproc_pipeline='vistasoft',\n",
    "    t1_preproc_pipeline='freesurfer',\n",
    "    tracking_params=tracking_params,\n",
    "    viz_backend_spec='plotly_no_mp4')\n",
    "\n",
    "fig_files = myafq.export_all()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b36f3f1f",
   "metadata": {},
   "source": [
    "## References\n",
    "\n",
    "[^Garyfallidis2017]: Garyfallidis, Eleftherios, Marc-Alexandre Côté,\n",
    "    Francois Rheault, Jasmeen Sidhu, Janice Hau, Laurent\n",
    "    Petit, David Fortin, Stephen Cunanne, and Maxime\n",
    "    Descoteaux. 2017. “Recognition of White Matter Bundles\n",
    "    Using Local and Global Streamline-Based Registration and\n",
    "    Clustering.” NeuroImage 170: 283-295.\n",
    "\n",
    ":::{only} html\n",
    "{download}`Download as Jupyter Notebook <recobundles.ipynb>`\n",
    ":::"
   ]
  }
 ],
 "metadata": {
  "file_format": "mystnb",
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "pygments_lexer": "ipython3"
  },
  "mystnb": {
   "execution_mode": "off"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}