notebooks/Collecting image statistics from Regions-of-Interest over many subjects.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import nibabel as nib\n",
    "from glob import glob\n",
    "import pandas as pd\n",
    "from joblib import Parallel, delayed\n",
    "\n",
    "def roistats_from_map(map_fp, atlas, func):\n",
    "     m = np.array(nib.load(map_fp).dataobj)\n",
    "     n_labels = list(np.unique(atlas))                                                                                                                                                                                                                                         \n",
    "     n_labels.remove(0)                                                                                                                                                                                                                                                        \n",
    "     label_values = [func(m[atlas==label]) for label in n_labels]                                                                                                                                                                                                              \n",
    "     return label_values                                                                                                                                                                                                                                                       \n",
    "                                                                                                                                                                                                                                                                               \n",
    "def roistats_from_maps(atlas_fp, maps_fp, subjects=None, labels=None,                                                                                                                                                                                                          \n",
    "        func=np.mean, n_jobs=7):                                                                                                                                                                                                                                               \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     atlas_im = nib.load(atlas_fp)                                                                                                                                                                                                                                             \n",
    "     atlas = np.array(atlas_im.dataobj)                                                                                                                                                                                                                                        \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     n_labels = list(np.unique(atlas))                                                                                                                                                                                                                                         \n",
    "     n_labels.remove(0)                                                                                                                                                                                                                                                        \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     df = Parallel(n_jobs=n_jobs, verbose=1)(\\                                                                                                                                                                                                                                 \n",
    "         delayed(roistats_from_map)(maps_fp[i], atlas, func)\\                                                                                                                                                                                                                  \n",
    "         for i in xrange(len(maps_fp)))                                                                                                                                                                                                                                        \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     columns = labels if not labels is None \\                                                                                                                                                                                                                                  \n",
    "             else [int(e) for e in n_labels]                                                                                                                                                                                                                                   \n",
    "     res = pd.DataFrame(df, columns=columns)                                                                                                                                                                                                                                   \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     res['subject'] = xrange(len(maps_fp)) if subjects is None else subjects                                                                                                                                                                                                   \n",
    "     res = res.set_index('subject')                                                                                                                                                                                                                                            \n",
    "                                                                                                                                                                                                                                                                               \n",
    "     return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=7)]: Done 137 tasks      | elapsed:    5.8s\n",
      "[Parallel(n_jobs=7)]: Done 505 out of 505 | elapsed:   16.2s finished\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Middle cerebellar peduncle</th>\n",
       "      <th>Pontine crossing tract</th>\n",
       "      <th>Genu of corpus callosum</th>\n",
       "      <th>Body of corpus callosum</th>\n",
       "      <th>Splenium of corpus callosum</th>\n",
       "      <th>Fornix</th>\n",
       "      <th>Corticospinal tract R</th>\n",
       "      <th>Corticospinal tract L</th>\n",
       "      <th>Medial lemniscus R</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subject</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10538</th>\n",
       "      <td>0.000817</td>\n",
       "      <td>0.000895</td>\n",
       "      <td>0.000757</td>\n",
       "      <td>0.000818</td>\n",
       "      <td>0.000833</td>\n",
       "      <td>0.000849</td>\n",
       "      <td>0.000819</td>\n",
       "      <td>0.000834</td>\n",
       "      <td>0.000716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12252</th>\n",
       "      <td>0.000850</td>\n",
       "      <td>0.000896</td>\n",
       "      <td>0.000719</td>\n",
       "      <td>0.000816</td>\n",
       "      <td>0.000817</td>\n",
       "      <td>0.000850</td>\n",
       "      <td>0.000843</td>\n",
       "      <td>0.000834</td>\n",
       "      <td>0.000743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55708</th>\n",
       "      <td>0.000853</td>\n",
       "      <td>0.000929</td>\n",
       "      <td>0.000732</td>\n",
       "      <td>0.000851</td>\n",
       "      <td>0.000795</td>\n",
       "      <td>0.000922</td>\n",
       "      <td>0.000839</td>\n",
       "      <td>0.000872</td>\n",
       "      <td>0.000786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12121</th>\n",
       "      <td>0.000842</td>\n",
       "      <td>0.000926</td>\n",
       "      <td>0.000810</td>\n",
       "      <td>0.000846</td>\n",
       "      <td>0.000826</td>\n",
       "      <td>0.000868</td>\n",
       "      <td>0.000810</td>\n",
       "      <td>0.000822</td>\n",
       "      <td>0.000723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11015</th>\n",
       "      <td>0.000847</td>\n",
       "      <td>0.000879</td>\n",
       "      <td>0.000805</td>\n",
       "      <td>0.000877</td>\n",
       "      <td>0.000856</td>\n",
       "      <td>0.000839</td>\n",
       "      <td>0.000814</td>\n",
       "      <td>0.000792</td>\n",
       "      <td>0.000831</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Middle cerebellar peduncle  Pontine crossing tract  \\\n",
       "subject                                                       \n",
       "10538                      0.000817                0.000895   \n",
       "12252                      0.000850                0.000896   \n",
       "55708                      0.000853                0.000929   \n",
       "12121                      0.000842                0.000926   \n",
       "11015                      0.000847                0.000879   \n",
       "\n",
       "         Genu of corpus callosum  Body of corpus callosum  \\\n",
       "subject                                                     \n",
       "10538                   0.000757                 0.000818   \n",
       "12252                   0.000719                 0.000816   \n",
       "55708                   0.000732                 0.000851   \n",
       "12121                   0.000810                 0.000846   \n",
       "11015                   0.000805                 0.000877   \n",
       "\n",
       "         Splenium of corpus callosum    Fornix  Corticospinal tract R  \\\n",
       "subject                                                                 \n",
       "10538                       0.000833  0.000849               0.000819   \n",
       "12252                       0.000817  0.000850               0.000843   \n",
       "55708                       0.000795  0.000922               0.000839   \n",
       "12121                       0.000826  0.000868               0.000810   \n",
       "11015                       0.000856  0.000839               0.000814   \n",
       "\n",
       "         Corticospinal tract L  Medial lemniscus R  \n",
       "subject                                             \n",
       "10538                 0.000834            0.000716  \n",
       "12252                 0.000834            0.000743  \n",
       "55708                 0.000872            0.000786  \n",
       "12121                 0.000822            0.000723  \n",
       "11015                 0.000792            0.000831  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "atlas_fp = '/home/grg/SPM/2016-2017/ROIapoE/ROI_DARTEL/csf5/rois.nii.gz'\n",
    "maps_fp = glob('/home/grg/data/ALFA_DWI/dartel_ALFA_DWI/dartel_csf.5/rswr*.nii')\n",
    "subjects = [e.split('/')[-1][4:9] for e in maps_fp]\n",
    "names = {1: 'Middle cerebellar peduncle', 2: 'Pontine crossing tract',\n",
    "        3: 'Genu of corpus callosum', 4: 'Body of corpus callosum',\n",
    "        5: 'Splenium of corpus callosum',  6: 'Fornix',\n",
    "        7: 'Corticospinal tract R',  8: 'Corticospinal tract L',\n",
    "        9: 'Medial lemniscus R'}\n",
    "df = roistats_from_maps(atlas_fp, maps_fp, subjects=subjects, labels=names.values())\n",
    "df.head()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11+"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}