levelset.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# General stuff\n",
    "import os\n",
    "import re\n",
    "import time\n",
    "import warnings\n",
    "import math\n",
    "import sys\n",
    "import scipy.io as sio\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import nibabel as nib\n",
    "from subprocess import Popen, PIPE\n",
    "from os.path import join as opj\n",
    "from IPython.core.debugger import Tracer\n",
    "from scipy.optimize import curve_fit\n",
    "\n",
    "# sklearn stuff\n",
    "from sklearn.utils import resample\n",
    "\n",
    "# personal functions\n",
    "from importlib import reload\n",
    "import fsutils as fs\n",
    "reload(fs)\n",
    "\n",
    "if os.path.exists('/data1/vbeliveau/'):\n",
    "    # On NRU server\n",
    "    cluster_code='/data1/vbeliveau/5HT_clustering'\n",
    "    cluster_data='/data1/vbeliveau/5HT_clustering_data'\n",
    "    subjects_dir='/usr/local/nru/freesurfer/subjects'\n",
    "elif os.path.exists('C:/Users/vbeliveau/'):\n",
    "    # On laptop\n",
    "    cluster_code='C:/Users/vbeliveau/Documents/GitHub/5HT_clustering'\n",
    "    cluster_data='C:/Users/vbeliveau/Documents/5HT_clustering_data'\n",
    "    subjects_dir='C:/Users/vbeliveau/Downloads/'\n",
    "else:\n",
    "    raise ValueError('Unknown location')\n",
    "    \n",
    "# Load local version of nilearn\n",
    "if os.path.exists('/data1/vbeliveau/'):\n",
    "    sys.path.append('/data1/vbeliveau/nilearn')\n",
    "    import mynilearn.decomposition as dcm \n",
    "    reload(dcm)\n",
    "elif os.path.exists('C:/Users/vbeliveau/'):\n",
    "    sys.path.append('C:/Users/vbeliveau/Downloads/nilearn-master')\n",
    "    import mynilearn.decomposition as dcm\n",
    "    reload(dcm)\n",
    "else:\n",
    "    raise ValueError('Unknown location')\n",
    "    \n",
    "# Analysis directories\n",
    "surf_data=opj(cluster_data,'surf_data')\n",
    "fs.assert_dir(surf_data)\n",
    "BPnd_data=opj(cluster_data,'BPnd')\n",
    "fs.assert_dir(BPnd_data)\n",
    "MFA_data=opj(cluster_data,'MFA')\n",
    "MFA_preproc=opj(MFA_data,'preproc')\n",
    "fs.assert_dir(MFA_data)\n",
    "fs.assert_dir(MFA_preproc)\n",
    "\n",
    "info_path=opj(cluster_code,'info_alltracers_base_healthy_hrrt.mat')\n",
    "hemi_type=['lh','rh']\n",
    "\n",
    "info=sio.loadmat(info_path)\n",
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "petID=[item for subl in info['petID'] for subsubl in subl for item in subsubl]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Run levelset on mean BPnd maps\n",
    "\n",
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "hemi_type=['lh']\n",
    "targ='fsaverage5'\n",
    "sm='10'\n",
    "\n",
    "data=np.empty(len(tracers),dtype=object)\n",
    "for tracer,nt in zip(tracers,np.arange(0,len(tracers))):\n",
    "    for hemi in hemi_type:\n",
    "        # Load data\n",
    "        fmask=opj(surf_data,'mask.'+targ+'.'+hemi)\n",
    "        fname=opj(BPnd_data,'maps',tracer,'mrtm2.nopvc.' + \n",
    "                  targ + '.' + hemi + '.sm' + sm + '.nii.gz')        \n",
    "        data=np.mean(fs.load_surf_data(fname,mask=fmask),axis=1)\n",
    "        \n",
    "        # Load neighborhood structure\n",
    "        fneigh=opj(surf_data,'neigh.cortex.'+targ+'.'+hemi+'.list')\n",
    "        neigh,_=fs.load_surf_neighborhood(fneigh)\n",
    "        \n",
    "        # Perform levelset clustering\n",
    "        clust=fs.surf_levelset(data,neigh)        \n",
    "        dest=opj(cluster_data,'levelset','test')\n",
    "        fs.assert_dir(dest)\n",
    "        fname=opj(dest,tracer+'.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(clust,fname,mask=fmask)\n",
    "        \n",
    "        # Perform levelset two-sided clustering\n",
    "        clust=fs.surf_levelset_two_sided(data,neigh)\n",
    "        dest=opj(cluster_data,'levelset','test.two-sided')\n",
    "        fs.assert_dir(dest)\n",
    "        fname=opj(dest,tracer+'.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(clust,fname,mask=fmask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/data1/vbeliveau/software/anaconda3/lib/python3.5/site-packages/nibabel/spatialimages.py:411: UserWarning: Using large vector Freesurfer hack; header will not be compatible with SPM or FSL\n",
      "  hdr.set_data_shape(shape)\n"
     ]
    }
   ],
   "source": [
    "# Bootstrap levelsets borders\n",
    "\n",
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "hemi_type=['lh']\n",
    "targ='fsaverage6'\n",
    "sm='10'\n",
    "maxiter=1000\n",
    "\n",
    "dest=opj(cluster_data,'levelset','test.two-sided.prob')\n",
    "fs.assert_dir(dest)\n",
    "\n",
    "for tracer,nt in zip(tracers,np.arange(0,len(tracers))):\n",
    "    for hemi in hemi_type:\n",
    "        # Load data\n",
    "        fmask=opj(surf_data,'mask.'+targ+'.'+hemi)\n",
    "        fname=opj(BPnd_data,'maps',tracer,'mrtm2.nopvc.' + \n",
    "                  targ + '.' + hemi + '.sm' + sm + '.nii.gz')        \n",
    "        data=fs.load_surf_data(fname,mask=fmask)\n",
    "        N=data.shape[1]\n",
    "        \n",
    "        # Load neighborhood structure\n",
    "        fneigh=opj(surf_data,'neigh.cortex.'+targ+'.'+hemi+'.list')\n",
    "        neigh,_=fs.load_surf_neighborhood(fneigh)\n",
    "        \n",
    "        clust_prob=np.zeros(data.shape[0])\n",
    "        for ni in np.arange(0,maxiter):        \n",
    "            # Perform levelset two-sided clustering\n",
    "            mdata=np.mean(data[:,resample(np.arange(0,N))],axis=1)\n",
    "            clust=fs.surf_levelset_two_sided(mdata,neigh)\n",
    "            clust_prob+=clust==-1        \n",
    "        clust_prob=clust_prob/maxiter\n",
    "            \n",
    "        fname=opj(dest,tracer+'.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(clust_prob,fname,mask=fmask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cumi, lh hemi\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhcAAAFkCAYAAACThxm6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDT+6AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb39e893470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dasb, lh hemi\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhcAAAFkCAYAAACThxm6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3Xd8VFX6x/HPQw2gIIpCEH4qq7BJUIGAfRWXtWBf1xa7\niIpiiwV1dRewgIAIiCKwKoIla8MVy4qVFUVUEhsJKgqoaACRKiQQkvP740wkhJmQMjczk3zfr9e8\nZnLOvTPPIeHmybmnmHMOERERkWhpEOsAREREpG5RciEiIiJRpeRCREREokrJhYiIiESVkgsRERGJ\nKiUXIiIiElVKLkRERCSqlFyIiIhIVCm5EBERkahSciEiIiJRFWhyYWZ/MrMZZvaTmZWY2SmVOKe3\nmWWbWaGZfWNmFwUZo4iIiERX0D0XLYDPgKuAHW5iYmZ7A68AbwMHAuOAR8zsmOBCFBERkWiy2tq4\nzMxKgNOcczMqOGYE0Nc5d0CZsiyglXPuhFoIU0RERGoo3sZcHAK8Va5sJnBoDGIRERGRamgU6wDK\naQcsL1e2HGhpZk2dc5vKn2BmuwHHAUuAwsAjFBERqTuSgL2Bmc65X6P1pvGWXFTHccBTsQ5CREQk\ngZ0HPB2tN4u35GIZ0LZcWVtgXbhei5AlAE8++SQpKSkBhhZ7mZmZjBkzJtZhVN2iRTBoEIwcCZ06\nhalexKBBgxg5ciSdQvUJ29YqUjvrlnhpZ3ExrFkD330HS5ZAfj78+CP88gssX+6fw0lKgkaNoGFD\n/2jQYOvXzZr5+gYNYMmSTLp0GfP7cY0aQYsWW88pfTRsCGbbljdqBI0b+9dm/lHV12bQpMn2ZeUf\nsO0xlfnazMfXsCGMHZvJ9deP2aau9Ln0dany71H2OdK5kerDfUabNv57EG0LFizg/PPPh9Dv0miJ\nt+TiQ6BvubJjQ+WRFAKkpKTQo0ePoOKKC61atUrcNi5e7BOLCPEvXryYTp06/d6+hG5rFaiddUtt\ntrO4GD7/HL7+GpYt88+LFsHKlfDFF74e/C/Kjh1hn32gVy9o187/N9x7b9h9961JQ7Nm/tjKOOWU\nVsyYUfe/ny+80IqLLqr77QyJ6rCCQJMLM2sB7AuU5l+dzOxAYJVz7kczGw60d86VrmUxERgYmjXy\nGNAHOAPQTBERqbeWLYO5c+GTT+CDD3wS8fPPWxOIZs3gD3+ALl1gr73goougQwdISYHOnX1vgUht\nCvpHrifwLn6NCweMDpVPBfrhB3B2LD3YObfEzE4ExgDXAkuBS51z5WeQiIjUWevXw8yZMGcOvPOO\n76EA3+tw2GFw+OE+eejaFbp1g5122r6bXiSWAk0unHP/o4Lprs65S8KUvQekBxmXiEitcg62bPGP\noqKtz0VFbNqwhQVfFPHZJ0V8NX8LixcW8dPiIijewp67F3FW2hZG37iFtD8W026PEigp8V0WJSXw\ncwn8GHpdtrw4TFllX5eU+HhL4y77KC3LzYXrr9++PNLxOyqr7nuE+3eOZtknn8Bf/xrsZ1S27O67\noXv37cvjlDrLEkhGRkasQ6g19aWtamcMOAebNsFvv219bNgQ/usNG6CwcPvHpk1hyzJ++cUPcCg9\npjSR2LIlYjhNgW6hx3Z+AWaFHlVROgKydFRlZV+XHzlZ+l7lHhmFhfDmm5FHUVY0wjLa54Rre2XK\nKnFsRrt2sHlz9N63JvEmWNeUkosEElcX6IDVl7aqndXknP/Fv2KFn/7w669+esTq1f4R6fXatT5p\nKCmp+P0bNPDTH1q02DrisWlT/1z2sdNOkJRESZMk1m5O4rBNSXy6MYmlvzRl1camrF7fmFXrG7Ns\nZSM2ucZsoRGtd29Mh30a06JVI3Zv35hddm/MPvs1Yu/9GtN0p8Zbp1OUPpdOXSg7haN8clD+64B/\nEdWPn9r6084gKLkQkfhRUuLnSf74o38sXernUZYmEStWbH1dULD9+U2bQuvWsMsu/rl1a9hzTz84\noXVraNXKJwTlHy1abPt1UtLvv6Cd8wMov/gC5s2D+fP9AMtff4V163wYGzduzVfMYL/94P/+D/ZI\ng+Rk6NbJz9bo3t2PmxCp65RcSPCSk2HwYP8ctjqZwYMHkxyhXuoQ53xy8O23Wx+LF29NJJYu3doN\nDX4xg+RkaNsW9tjDJwmlr/fYY+vr3XbzyUMNFgLYsgV++AGWfAnff+9DmjULcnJ8hwf4tQbS030Y\nu+3mc5hmzaB5c9h3X39HZM89fa4iUp/V2sZlQTGzHkB2dnZ2vZhHL5IQiorgm2/8n/nz5/tFGEqT\nifXrtx63557+T/qOHf2jQ4etrzt29AsxRLGLf8MG3+uwdKmfyrl0KXz1Fbz3nn9dWGam/667whFH\n+LUh0tPhgAN8uCJ1SU5ODunp6QDpzrmcaL2vei5EpGZWrfL3C7KzfSLx5Zf+N3ZRka9PTvYLLvTq\nBRkZ/k/8/fbzi6o1bx7VUEpKfK9Dbq7PZ7780vdGLF3q76SsW7ft8Tvv7NeHOO44H1JKil8nYp99\nfKeJiFSPkgsRqbxNm3wi8ckn8PHH/vnbb31dq1aw//5+EYYrrvCv09L8/YMoWrMGPvoI8vJ8PrNm\njU8cVq/2vRIbNvjjmjXzIey1l1+Rsl07fxclOdl3irRr50MWkehTciEikRUU+N/ks2bB//7nl4ks\nLPQDHrt3hxNP9D0SBx3kuwAaRFzWpsrWrfMJRH6+z1/y8vzHf/ed7xRp0gR69vTjIA480Ocwu+/u\nV6ns3h3at49qOCJSBUouRGSr0gWSXnsN/vtfv0Tk5s1+sOSRR8KwYf75gAMqvxFFJT5y+XJ4/31/\nG+Pll/0YzzVrth6z007+tsWf/wzXXgvHHON7H5o2jUoIIhJlSi5E6rvCQnjrLXj1VZ9U/PCDv6fQ\np4/fybZ3b39/IQrdACtW+Cmdb7/tZ2EsXOiHbJTOxmjd2o9/OPts3/PQtasfRBnlcZ0iEjAlFyL1\n0aZN8MYb8Oyz8NJLfgbHvvvCaafBCSfAUUf5Wx81VFDgeyReeQXefdf3TIAf79CzJ5x5pp+V8Yc/\n+D0z2rev8UeKSBxQciHBKyjwqxB16hR2HYKCggIWLVpEp06daFaDdQpkB5zzcy6nTIEXX/SDGtLS\n4MYb/W/51NSofMzKlT5vefVV/1i71s8w/fOf4bbb/McccIB6IkTqMiUXErwFC/xCAdnZftj+dtUL\nSE9PR2uVBGTpUpg61ScV333nuwkyM31CkZYWlY/49FN48kmfTHz9tS/r2hWuu27rxyiZEKk/lFyI\n1EXO+b26x4714yiSkvxv+ccegz/9KSq/6Rcu9Lc6srL8ZJK2bf3kkSFD/EdowSmR+kvJhUhdUlAA\nTz0F48b5Ba0OOAAmTvQjJFu2rPHbL13qZ3NMmODf3sxPHnnmGTj9dL/XloiILgUidcG6dfDQQ3D/\n/X5HrVNOgQce8DM9atBLsXYtvP46zJ7tB2V+/72fNHLSSXDnnX4chRaiEpHylFyIJLI1a3wSMXas\nX5qyf3+44QY/rqIGvv3Wv+XUqX6H8r33hlNP9UNnTjrJz/AQEYlEyYVIItq4EcaMgVGj/LTSK66A\nm2+u0UCH/HyYPBlmzvSLcrZqBddcAwMG+O3DRUQqS8mFSCIpLoZp0+Af//ArUl11Fdx6q184ohqc\n88tcTJ4Mb77pl9Q++WS48EK4+OKoLHUhIvWQkguRRPHuu3D99X6Jy7PO8ktxV/P2R1GRH/c5YoTf\nwLR7dxg9Gi64wK+SKSJSE0ouJHgpKX5qQadOEapTmD9/Pp0i1Nd7y5f7ha6eesovYzl3Lhx8cLXe\nasMGePppuOsu+PFH+MtfYPx4/ywiEi1KLiR4zZpVuFhTs2bNSIvSYk51SnGxv19x221+judjj8FF\nF1V5j4+1a/04iuef97dANm/2M1PvuMMvdCUiEm1KLkTi0bff+kEPH3wAl17q71/stluV3mLjRnj4\nYbjnHli92ncc3XOPn6XauXMwYYuIgJILkfjinM8Ibr7ZD9L83//8KlVVUFLiT3/oIdiyxc9O/fvf\nNeNDRGqPkguRePHTT3DJJX7axpVX+u3Od9qp0qeX7kt2xx2+w+Of//SzPjSURURqm5ILkXjw+ut+\nqkbTpn6AxLHHVun0Tz7xYz5nz/YzP156yU8pFRGJhaqNDBOR6NqyBW6/Hfr2hV694LPPqpRYFBfD\n3XfDQQf5cRUzZsC8eUosRCS21HMhEivLlvlpGx98AMOHw6BBlZ4JsmmTP23QIMjJ8bdA/vlPaNgw\n4JhFRCpBPRcSvPx8vw93fn6E6nyGDBlCfoT6Oiknx/dUlO5bfuutlUosior8QM0OHaBPH7/vx+zZ\nMHSoEgsRiR9KLiR4+fn+t18FycXQoUPrT3Lx3HNwxBF+Nsgnn8Cf/lSp0374wY+nuPpqf9vj008h\nLw8OPzzgeEVEqkjJhUhtcc4nWWed5bcYfe+9Sm809t57sP/+sH49vP++X0+rW7cqr6clIlIrdGkS\nqQ1btvjFsIYM8WtvP/20X7m0EqeNG+fHePbq5e+mqKdCROKdBnSKBG3jRj9w8/XX4Ykn4PzzK3Xa\nBx/4TU+//BIuv9xvLNaiRcCxiohEgXouRIK0ahUccwy88w68/HKlEovFiyEjww/LSErywzImTlRi\nISKJQz0XIkFZvtxvN5qf75OLHexkWlgIo0b5ndR3283vWXbppRpXISKJR8mFSBCWLYM//9mvbDV7\ntt92vgKzZ0O/frBkCdxwA/zjH1Va+VtEJK7obyIJXlISpKb657DVSaSmppIUoT7h5OfD0UfDmjUw\na1aFicX69T6p6N3bz0z9/HO/AaoSCxFJZOq5kOClpkJubgXVqeRWUJ9Qfv7ZJxYbNvgdTffbL+Kh\n//ufH4KxejWMHQsDBkDjxrUYq4hIQNRzIRItK1f6MRYbN/oeiwiJxVtv+fyjd2/Yd1/44gu45hol\nFiJSdyi5EImG9evhhBN8gvHWWz5rKCcvDw491E8eWb0annzS766uLdFFpK5RciFSU4WFcNpp8PXX\nfi2LLl22qS4q8jNAunf3wzBeeMEv3X3eedBINyZFpA4KPLkws4FmttjMCsxsrpn12sHx55nZZ2a2\nwcx+NrNHzWzXoOMUqZYtW+Dcc2HOHL+ORY8e21R//jkccoif/ZGZ6ZOK008HsxjFKyJSCwJNLszs\nbGA0MBjoDnwOzDSzNhGOPxyYCvwLSAXOAA4CJgcZp0i1OAfXXQczZsDzz8ORR/5eVVjotxHp2RM2\nb4a5c+HeeyNOmBERqVOC7rnIBCY556Y5574CBgAbgX4Rjj8EWOyce8g5971zbg4wCZ9giMSXceNg\nwgS/fOaJJ/5evGABHHCA30Lklltg3jy/L4iISH0RWHJhZo2BdODt0jLnnAPeAg6NcNqHQEcz6xt6\nj7bAmcCrQcUpUi0zZvjVrgYNgv79ASguhocfhvR0P5biyy/h7ruhadMYxyoiUsuC7LloAzQElpcr\nXw60C3dCqKfifOAZM9sM5AOrgasDjFOClpcHaWn+OWx1HmlpaeRFqI872dl+84/TT4fhw3EO/vtf\nvwX6VVf5tSvmzdvhopwiInVWXI1VN7NUYBwwBHgDSAbuw98a6V/RuZmZmbRq1WqbsoyMDDIyMgKJ\nVaqgsNAnFoWFEaoLycvLozBCfVzJz4dTToGuXWHaNL75tgHXX++TiyOPhI8/1i0QEYlPWVlZZGVl\nbVO2du3aQD4ryORiJVAMtC1X3hZYFuGcW4EPnHP3h76eb2ZXAbPN7HbnXPlekN+NGTOGHuVG6otE\n1ebNcMYZ/vVLL/Hk9OZceim0bw/Tp/vZqJoFIiLxKtwf3Dk5OaSnp0f9swK7LeKcKwKygT6lZWZm\noa/nRDitObClXFkJ4ABdtiW2MjPhk0/YnPUCl97ejgsu8AlFXh789a9KLERESgV9W+R+4HEzywY+\nxs8eaQ48DmBmw4H2zrmLQse/DEw2swHATKA9MAb4yDkXqbdDJHiPPw4TJrDy7on0vuoQvvkGxozx\nM1GVVIiIbCvQ5MI592xoTYs78bdDPgOOc879EjqkHdCxzPFTzWwnYCB+rMUa/GyTW4OMU6RC2dkw\nYAA/HX8p6Q9cTvMWvmj//WMdmIhIfAp8QKdzbgIwIULdJWHKHgIeCjoukUpZswZ3xhn8uMsBdH79\nQY48xnjiCWhbfiSRiIj8Lq5mi4jEFedYf3Z/3I9r6F3yLvfcl0RmJjTQjjwiIhVSciHBS06GwYP9\nc9jqZAYPHkxyhPpYWXD9JFLeeIGrk19g2rN7c8QRsY5IRCQxmF80M3GZWQ8gOzs7W1NRJSqcg6k3\nfsE5Yw7ijY6X8pevHqJ581hHJSISfWWmoqY753Ki9b7q4BUpo6QEbrt2AweNOYdVbbpwzOejlViI\niFSRbouIhGzeDAMGwCFTMtmvyfc0fm8etNY2piIiVaXkQgRYswZOOgnafPQql/MveGCiNgcREakm\nJRdS7y1fDiecAKsW/srbLfvDwX3h8stjHZaISMLSmAup1xYtgj/9CX76CeYdMpCmbhM88oiW3RQR\nqQElF1JvvfMOHHggFBfD57f9m93efAYmTPA7kYmISLUpuZDgFRRAbq5/DltdQG5uLgUR6oPw3//C\nySfDIYdA9ss/03boVXDWWXDOObUWg4hIXaXkQoK3YAF07eqfw1YvoGvXriyIUB9t778Pp58OffrA\nS/9x7HLbldCkie+1EBGRGtOATqlX5s3z26MfdBA89xw0ffkFmDEDnn8edtst1uGJiNQJ6rmQemPS\nJD948w9/gOnToenG1XDNNXDaab4rQ0REokLJhdQLjzziF8g6/3x4991QJ8WgQbBxIzz4oGaHiIhE\nkW6LSJ03ebJPLAYOhPHjQ3nErFk+43j4Ydhzz1iHKCJSp6jnQuq0uXPhiivgkkvggQdCiUVBgV8k\n64gjtFiWiEgA1HMhddaGDT6p6N7d9140KE2lhw2D77/3AzkbKL8WEYk2JRdSZ91wg88hcnKgYcNQ\n4cKFMHIk3HIL/PGPMY1PRKSuUnIhwUtJgfnzoVOnCNUpzJ8/n04R6qvjww99b8WYMWVyCOfg2mv9\nCpy33Ra1zxIRkW0puZDgNWsGaWkVVDcjrYL6qlq+HM49F3r2hKuvLlPx0kvw+uvwn//4mEREJBC6\n4Sx1yooVfi2LzZv9IlmNStPnjRvh+uuhb1845ZSYxigiUtep50LqjE2b4OyzYfVqf1tk773LVN57\nL+Tnw1tvaU0LEZGAKbmQOsE5uPBCn1S8/jrsu2+Zym+/hREj4Oaby1WIiEgQlFxInfDkk/Dss36L\nkN69y1VmZkK7dvD3v8ciNBGRekfJhSS8/HyfP5x9Nvztb+Uq33wTXnnFZx7Nm8ckPhGR+kYDOiWh\nFRX58ZkNG/qlvbdRXAw33giHHw5nnBGT+ERE6iMlFxK8/HwYMsQ/h63OZ8iQIeRHqK/IhAl+kaxX\nXoHddy9XOWUKfPkljB6tQZwiIrVIyYUELz8fhg6tMLkYOnRolZOLxYvhjjv89iC9epWrXL8e/vEP\nv+DFwQdXM3AREakOJReSkJyDfv1g1139RJDtjBwJa9bA8OG1HpuISH2nAZ2SkP7zH79r+syZ0LJl\nucoff4T77vObi/zf/8UiPBGRek09F5JwNmzwW4Qcdxwce2yYA/7+d59x3HprrccmIiLquZAEdO+9\n8MsvfjDndrKz/aIXkybBzjvXemwiIqKeC0kwCxb4YRQ33hhhk9XbbvO7sPbrV+uxiYiIp54LSRib\nN8N55/kVvO+4I8wBb7/tF82aPr3MjmUiIlLbdAWW4CUlQWqqfw5bnURqaipJEepLZWb6ZSs++ijM\njunO+V6Lgw6C006LUuAiIlIdSi4keKmpkJtbQXUquRXUg58AMnky3H039OgR5oAXX4RPPoF33tGC\nWSIiMaYxF5IQBg2C1q1hwIAwlVu2wO23+6kjRx9d67GJiMi21HMhce+jj+Df/4apU6FVqzAHTJsG\nX33lZ4mIiEjMqedC4t6IEdC5sx/MuZ3CQhg8GM46C9LTaz02ERHZnnouJK7Nn++HUzzyiN/5dDsT\nJvg9S+66q9ZjExGR8ALvuTCzgWa22MwKzGyumZXfYqr88U3M7B4zW2JmhWa2yMwuDjpOiU/DhvkV\nvC+4IEzl+vX+gH79fNeGiIjEhUB7LszsbGA0cDnwMZAJzDSzzs65lRFOew7YHbgE+A5IRrdv6qWF\nC+GZZ2D8eGjSJMwBDz64dfdTERGJG0HfFskEJjnnpgGY2QDgRKAfMLL8wWZ2PPAnoJNzbk2o+IeA\nY5Q4dffd0LZthMU2163zm5P17w8dO9Z6bCIiEllgPQJm1hhIB94uLXPOOeAt4NAIp50MzANuMbOl\nZva1mY0ys4pXV5L4lpcHaWn+OWx1HmlpaeSVqV+0yE/+uO22CGtvjR8Pv/3mDxARkbgSZM9FG6Ah\nsLxc+XKgS4RzOuF7LgqB00Lv8TCwK3BpMGFK4AoLfWJRWBihupC8vDwKy9Q/+CDssovvmNjO2rUw\nejRcdhl06BBQ0CIiUl3xNlukAVACnOuc+w3AzG4AnjOzq5xzm2IandSK336DRx+Fq64Ks8w3+F6L\njRvVayEiEqeCTC5WAsVA23LlbYFlEc7JB34qTSxCFgAGdMAP8AwrMzOTVuVWWMrIyCAjI6OKYUus\nPfusTzCuvDJMZWmvxeWXw5571npsIiKJKisri6ysrG3K1q5dG8hnBZZcOOeKzCwb6APMADAzC339\nQITTPgDOMLPmzrmNobIu+N6MpRV93pgxY+gRdtMJSTRTpkDv3n4K6nbGjfO3V269tbbDEhFJaOH+\n4M7JySE9gAUIg57ieT9wmZldaGZ/BCYCzYHHAcxsuJlNLXP808CvwBQzSzGzI/GzSh7VLZH6ITcX\n3n/f3xLZzpo1cP/9cMUV0L59rccmIiKVE+iYC+fcs2bWBrgTfzvkM+A459wvoUPaAR3LHL/BzI4B\nxgOf4BONZwAtZFBPPPcctGwJJ50UpnLsWNi0CW65pdbjEhGRygt8QKdzbgIwIULdJWHKvgGOCzou\niU8vvACnnAJNm5arWLPGJxdXXgnJyTGJTUREKkcrX0rwkpP95mIRkoLk5GQGDx7Mhg3JzJ8Pf/tb\nmIPGj/djLQYNCjZWERGpMSUXErzkZBgypMLkYsiQIcyenUyLFnBc+X6r337zvRb9+0O7doGHKyIi\nNaPkQuLG88/7sRbbrW0xaZJf7lu9FiIiCUHJhcSFRYvg00/D3BIpLPR7iFxwQYS5qSIiEm+UXEhc\neOEF32PRt2+5iilTYMUKrWshIpJAlFxIXJgxA449FnbaqUxhURGMGAFnngmdO8csNhERqZp421tE\n6qEffoAPPvD7iWzj6afh++995iEiIglDPRcSc9OnQ+PGcMYZZQqLi2H4cDj5ZDjggJjFJiIiVafk\nQoJXUODX9S4oCFudlVXAYYfl0qhRmfoXX4Svv4bbb6+lIEVEJFqUXEjwFiyArl39czlLlsDHHy9g\n1qyuLCitdw6GDYM+feDgg2s3VhERqTGNuZCYev55aNIENm8uU/jf//p5qe+8E7O4RESk+tRzITH1\n6qtw0EFlCpyDe+6BQw/1+66LiEjCUXIhMbNiBbz3Hhx1VJnC//0P5szxYy3MYhabiIhUn5ILiZmX\nXvL5w9FHlykcOdLPDjnhhJjFJSIiNaMxFxIz77wDPXtC69ahgoUL/XiLJ55Qr4WISAJTz4XERGEh\nvPZauR1Qp02Djh3h7LNjFpeIiNSckguJidde8xudnnNOmcKZM+GGG/yKWiIikrCUXEjwUlJg/nz/\nHDJtmr8lkpICKSkpzL/wQlJ23hn6949hoCIiEg1KLiR4zZpBWpp/BjZtgtdfh7POClUXFpI2fTrN\nBg4st3OZiIgkIiUXUuvefdcnGL+Pt5g40e+Aes01MY1LRESiQ8mF1LrnnoP99oP998eP7Bw3Di66\nCNq2jXVoIiISBUoupFYVFfk9yc48MzTb9Ikn/GpaN94Y69BERCRKlFxIrXr7bVi92icXlJTAfffB\nX/8KnTvHOjQREYkSLaIlteq55+APf4ADDwRemgHffOOnjoiISJ2hngupNUVF8J//hG6J4GDECDjy\nSG2rLiJSxyi5kODl58OQIcx+Np9Vq0JTUD/4AObOhZtvJj8/nyFDhpCfnx/rSEVEJAqUXEjw8vNh\n6FBmPp5P167QrRt+g7LUVDjhBPLz8xk6dKiSCxGROkJjLqTWvDsLLhkBtiAPXn4ZpkyBBspvRUTq\nGl3ZpdY0bAAXXwyMHg3t28O558Y6JBERCYCSCwmcc/65Tx/YtfBnv7ZFZiY0aRLbwEREJBBKLiRw\n8+f755NPxq/G2awZXH55TGMSEZHgKLmQwL3zjn/u+cf1fh+RAQOgZcvYBiUiIoFRciGBe/99/9zw\nP9OhoACuuy62AYmISKCUXEigli6F3EVJrN0zBbKy4IIL/GDOMpKSkkhNTSUpKSlGUYqISDQpuZBA\nvfoqfGWpNLx1EPzyC9x003bHpKamkpubS2pqagwiFBGRaFNyIYGaOhVOOL6EnR4eBaecAikpsQ5J\nREQCpuRCApOfDx9+CJldXoO8PLj55liHJCIitUDJhQTm1Vf9ApxHzh0Jhx4Khx8e65BERKQWaPlv\nCcyMGXBp2lwaz50NL74IZrEOSUREaoGSCwnEihXw2mvw9f6joHNnP95CRETqBSUXEojnnoPOfEOn\nz1+ESZO0QZmISD0S+BXfzAaa2WIzKzCzuWbWq5LnHW5mRWaWE3SMEn1PPw2j24/G9tjDr20hIiL1\nRqDJhZmdDYwGBgPdgc+BmWbWZgfntQKmAm8FGZ8E44cf4Ns5yzl22VS/GueiRZCW5meMhJGXl0da\nWhp5EepFRCSxBN1zkQlMcs5Nc859BQwANgL9dnDeROApYG7A8UkAnn8eMhuOp0HTxn4fkcJCn1gU\nFoY9vrCwkLy8PAoj1IuISGIJLLkws8ZAOvB2aZlzzuF7Iw6t4LxLgH2AoUHFJsGaPu03BjaYgF12\nGbRuHesWcfjZAAAbcklEQVRwRESklgU5oLMN0BBYXq58OdAl3Almth8wDDjCOVdimrqYcD79FNI/\nf5QWDdfD9dfHOhwREYmBuJktYmYN8LdCBjvnvistruz5mZmZtGrVapuyjIwMMjIyohek7NATjxVx\nc4P74ZwM+L//i3U4IiISkpWVRVZW1jZla9euDeSzgkwuVgLFQNty5W2BZWGO3xnoCXQzs4dCZQ0A\nM7PNwLHOuVmRPmzMmDH06NGjxkFL9RUVQcHUZ+lQ8gMM2n6DMhERiZ1wf3Dn5OSQnp4e9c8KbMyF\nc64IyAb6lJaZv8/RB5gT5pR1QFegG3Bg6DER+Cr0+qOgYpXoeGOmY8D6kaw7/Hg44IBYhyMiIjES\n9G2R+4HHzSwb+Bg/e6Q58DiAmQ0H2jvnLgoN9txmLqKZrQAKnXMLAo5TouCzUW9yIl/AXWNjHYqI\niMRQoMmFc+7Z0JoWd+Jvh3wGHOec+yV0SDugY5AxSO1YswYOnT2SZR3Sade797aVyckweLB/DiM5\nOZnBgweTHKFeREQSi/kOg8RlZj2A7OzsbI25iKEXbs/hb8PSWT3xGVpfcVaswxERkUooM+Yi3TkX\ntRWxteGD1Jhz0OLhUeQ370TrS0+PdTgiIhJjSi6kxnJfWcwxq59l5YU3QKO4md0sIiIxouRCauzX\nO+5nrbUmZcQlsQ5FRETigJILqZGi/JX0+uJR5h16DY1aNo91OCIiEgeUXEiNfHfTBAA6DB8Y40hE\nRCReKLmQ6tu4kfbPj2fGbv1IPbJNrKMREZE4oeRCqq1g4lRabF7Fb5ffsIMDCyA31z+HrS4gNzeX\nggj1IiKSWJRcSPVs2ULxiFE8zxkcc0Wnio9dsAC6dvXPYasX0LVrVxZEqBcRkcSi5EKq55ln2GnF\nYl7d/zb22ivWwYiISDzRogRSdSUlbB46nLfoy19u6hbraEREJM4ouZCqe+UVmizMZWyzibz4t1gH\nIyIi8UbJhVSNc7hhw5jX9Aj+79wjaNEi1gGJiEi8UXIhVTNrFvbRR/yT17j94lgHIyIi8UjJhVTN\n8OF8v2s3Fu5yPIcfHutgREQkHim5kMqbNw/efJM7mjzDxZmGWawDEhGReKTkQipv+HBW774fz/z6\nNxZdXIXzUlJg/nzoFH49jJSUFObPn0+nCPUiIpJYlFxI5SxYANOnM6LNI5yV0ZAOHapwbrNmkJZW\nQXUz0iqoFxGRxKLkQipnxAgKdt2T+1dewPvXxjoYERGJZ0ouZMe+/x731FNMaj+Sw/ZvwkEHxTog\nERGJZ0ouZMfuu4+i5q2444fLeG5irIMREZF4p71FpGL5+fCvf/F0m+v4wwE7cfzxsQ5IRETinXou\npGIjR7KlcRLXLbqWfz2Dpp+KiMgOKbmQyJYtg4kTeabDLezZuBV/0z4iIiJSCbotIpHddx9FDZpw\n9bfX8c9/QsOG1Xyf/HwYMsQ/h63OZ8iQIeRHqBcRkcSi5ELCW7ECJkzg2XbX0qlHa84+uwbvlZ8P\nQ4dWmFwMHTpUyYWISB2h2yIS3ujRlDRoyLWLrmfsNI21EBGRylPPhWxv5Up46CHm9LiGDU1349RT\nYx2QiIgkEiUXsr3778cB/ebfwGWXQcuWsQ5IREQSiZIL2davv8L48Xx22EC+XdOGa66JdUAiIpJo\nlFzItsaOxZWUcGnejZxzDnTuHOuAREQk0Si5kK1+/RUeeICFfQbw6U97MGhQrAMSEZFEpORCtho1\nCoqLuW31LRx+OHTrFqX3TUqC1FT/HLY6idTUVJIi1IuISGLRVFTxli2DBx5g5QWZTJ+8B089FcX3\nTk2F3NwKqlPJraBeREQSi3ouxBs2DJo0YZS7iT32QEt9i4hItSm5EPjhB5g0id+uvJkJWa3p3x+a\nNo11UCIikqiUXAjcdRe0bMlYdx1mcMMNsQ5IREQSmZKL+m7hQpgyhc033saDj+9ERgbstlusgxIR\nkUSm5KK+GzIE2rZlWosr+eUXuOWWWAckIiKJTrNF6rP58yEri5IHJ3D3yGacdBJ06hTroEREJNEp\nuajP7rgD9t6b2fv14/vv4fHHYx2QiIjUBbotUl+9/z689BLcdRcTH2tCWhocdVRAn5WXB2lp/jls\ndR5paWnkRagXEZHEEnhyYWYDzWyxmRWY2Vwz61XBsX81szfMbIWZrTWzOWZ2bNAx1jvOwc03Q48e\nrD4+g5dfhnPOAbOAPq+w0CcWhYURqgvJy8ujMEK9iIgklkCTCzM7GxgNDAa6A58DM82sTYRTjgTe\nAPoCPYB3gZfN7MAg46x3pk+HuXNh1CgenNCALVvgsstiHZSIiNQVQY+5yAQmOeemAZjZAOBEoB8w\nsvzBzrnMckW3m9mpwMn4xERqqqgIbr0V+val8LA/M+4s6N8f2raNdWAiIlJXBJZcmFljIB0YVlrm\nnHNm9hZwaCXfw4CdgVWBBFkfTZoEixbBiy/yn//4jVCvuSbWQYmISF0S5G2RNkBDYHm58uVAu0q+\nx81AC+DZKMZVf61bB0OHwsUXQ9euPPooHHYYdOkS68BERKQuidupqGZ2LvAP4BTn3ModHZ+ZmUmr\nVq22KcvIyCAjIyOgCBPQyJGwYQMMHcpPP8Hbb8O//hXroEREpDZkZWWRlZW1TdnatWsD+awgk4uV\nQDFQ/m5+W2BZRSea2TnAZOAM59y7lfmwMWPG0KNHj+rEWT8sXQr33w+ZmdChA//oB61bw+mnxzow\nERGpDeH+4M7JySE9PT3qnxXYbRHnXBGQDfQpLQuNoegDzIl0npllAI8C5zjnXg8qvnpn0CBo2RJu\nuYXsbJgyBe65xycYgUtOhsGD/XPY6mQGDx5McoR6ERFJLEHfFrkfeNzMsoGP8bNHmgOPA5jZcKC9\nc+6i0NfnhuquBT4xs9JejwLn3LqAY6273n8fsrJ8RtGyJWPG+GW+a236aXKy38MkYnUyQyqoFxGR\nxBLoOhfOuWeBm4A7gU+BA4DjnHO/hA5pB3Qsc8pl+EGgDwE/l3mMDTLOOq24GK67Dnr1ggsvZMkS\n+Pe//QyRhg1jHZyIiNRFgQ/odM5NACZEqLuk3NdHBx1PvTNlCuTkwJw50KABDz4Iu+yiRbNERCQ4\n2lukLlu7Fv7+dzj/fDj0UIqL4amn4LzzoEWLWAcnIiJ1lZKLuuzOO2HjRrj3XgDefBOWLYMLLohx\nXCIiUqcpuairvvoKHnjA91zsuScAEydCt24QwKwjERGR3ym5qIucgyuvhL32ghtuAPwyF6+8Apde\nGuDupyIiIii5qJuefBJmzYIJEyApCYCHH/YvY3JLpKAAcnP9c9jqAnJzcymIUC8iIolFyUVds2oV\n3HgjnHMOHHssAD/9BOPG+V6Lciuk144FC6BrV/8ctnoBXbt2ZUGEehERSSxKLuqa226DTZv8Ut8h\nN97oZ4cMHRrDuEREpN6I243LpBrmzIHJk+HBB39fajsnB555Bh57zK9vISIiEjT1XNQVRUV+EGfP\nnjBgwO/Fw4bBvvvChRfGMDYREalX1HNRV9x3H8yfDx9//Pu63p9/DtOnw6RJWupbRERqj3ou6oK8\nPL8x2E03bbOIxfDhsPfecNFFMYtMRETqIfVcJLriYujXD/bZZ5udR2fN8mMtJk+GJk1iFp2IiNRD\nSi4S3dix/lbI++9Ds2YAlJTArbf6Toz+/WMcn4iI1DtKLhLZwoVwxx1+S/XDDvu9+N//ho8+gnfe\niZPVOFNS/HiQTp0iVKcwf/58OkWoFxGRxKLkIlGVlPhVsdq3h7vv/r140ya/ncipp8LR8bKBfbNm\nkJZWQXUz0iqoFxGRxKLkIlGNHu1vhbzzzjb7pz/4oN9HZObMGMYmIiL1mmaLJKLPPoPbb/dLb/bu\n/XtxUZGfkdqvH3TpErvwRESkflNykWgKCuC88yA1dZvbIQDTpsGyZX4IhoiISKzotkiiueUW+O47\nyM6Gpk1/Ly4qgnvugTPOqHB4g4iISOCUXCSS11+H8eP9FqflMojHH4fFi+Gll2ITmoiISCndFkkU\ny5bBxRfDccfB1VdvU7Vqld8M9bzzYP/9YxOeiIhIKSUXiaC4GDIy/KIVU6dCg22/bUOHwubNfgJJ\nXMrP96uH5udHqM5nyJAh5EeoFxGRxKLkIhEMGQLvvedXx2rbdpuqxYvhoYf85JFyVfEjP99nQBUk\nF0OHDlVyISJSR2jMRbx7/XU/K2TYMDjqqO2q77kHdtlluzslIiIiMaPkIp79+COcfz707etniZTz\n3Xd+IOd9922zjpaIiEhM6bZIvCoshDPP9EtnT5u23TgL8Hca9tgDrrgiBvGJiIhEoJ6LeOSczxg+\n/xxmz4Y2bbY7ZM4ceOIJePjh3zdDFRERiQtKLuLR2LG+t+Kpp6Bnz+2qS0pgwADo1QsuuywG8YmI\niFRAyUW8eeMNuOkmGDQIzj037CGPPAJffgkffggNG9ZyfCIiIjugMRfx5Jtv4Oyz/UJZw4aFPWT1\nar+l+oUXwiGH1HJ81ZWU5PdCSUqKUJ1EamoqSRHqRUQksajnIl4sXw7HHw/JyfD00xG7JCZPhnXr\n4N57azm+mkhNhdzcCqpTya2gXkREEouSi3iwYQOcdJLf8fTdd/3CFWGUlPhbIuec43MQERGReKTk\nIta2bPG3Qr76yq/CuddeEQ+dPh2+/daP9RQREYlXSi5iyTkYOBBmzoRXX4Xu3SMeWlICd90FxxwD\nhx5aizGKiIhUkZKLWHHOzwiZPBmmTIFjj63w8Kefhi++8MteiIiIxDPNFomVoUP9ut3jxvmt1Cuw\nbh1cd53fGPWII2onPBERkepSchELo0b55GLYMLj22h0ePn48rFnjTxMREYl3Si5q2/jx/nbI7bfD\nbbft8PAVK3xSkZEBe+5ZC/GJiIjUkJKL2nTffb6n4sYb/ejMHVi/Hk45xe9ZNnZsLcQXlLw8SEvz\nz2Gr80hLSyMvQr2IiCQWJRe1wTm48064+WbfYzFqFJhVeEpREZx4ov99/MYbYfcuSxyFhb4hhYUR\nqgvJy8ujMEK9iIgklsCTCzMbaGaLzazAzOaaWa8dHN/bzLLNrNDMvjGzi4KOMVDO+dsfgwfDPffA\n3XdXKrE44wy/8+l//xt27zIREZG4FWhyYWZnA6OBwUB34HNgppmF/TvczPYGXgHeBg4ExgGPmNkx\nQcYZmC1b/NbpI0bAmDF+U5Ad+PVXOPlkeO01mDEDDj+8FuIUERGJoqB7LjKBSc65ac65r4ABwEag\nX4TjrwQWOecGOee+ds49BDwfep/E8ttvfsDElCnw+ONw/fU7POXll/06WvPm+TW1Tjgh+DBFRESi\nLbDkwswaA+n4XggAnHMOeAuItMbkIaH6smZWcHx8WrYMjjoK3n/fd0FcVPGdnUWLoH9/n4ukpUF2\n9g7X1BIREYlbQa7Q2QZoCCwvV74c6BLhnHYRjm9pZk2dc5uiG6IfElFSsu2jtKyoCDZv9l+XPkrr\nI5U1zvuc9ledCkVF/Dh1NoV7HojLDX/uDz/4jo1XXoHWrf0s1YEDdzgkQ0REJK7V+eW///IX+Oij\n7ROI0kc0ncUzTOES5tOFU3mJH0//vx2e88c/+mmm/fpB8+bRjUdERCQWgkwuVgLFQNty5W2BZRHO\nWRbh+HU76rXIzMykVatW25RlZGTQv38GJ53k14po0MD3CpS+jvQoPaZhQ0hK8l+XPkrrt/m6pJg/\nPPZ39vr3SJb/5Vw2DvoXTzdrXuF5ZrD77tCxYz3oqUhO9rNlIuwTn5yczODBg0nWPvIiIoHJysoi\nKytrm7K1a9cG8lnmh0EEw8zmAh85564LfW3AD8ADzrntFrM2s3uBvs65A8uUPQ3s4pwLO7zRzHoA\n2dnZ2fTo0SOIZlRs2TK48EJ4+20YORJuuKEeZAsiIlIX5OTkkJ6eDpDunMuJ1vsGfVvkfuBxM8sG\nPsbP+mgOPA5gZsOB9s650hGPE4GBZjYCeAzoA5wBxOe8iTfegAsu8K9nzvT3YEREROq5QKeiOuee\nBW4C7gQ+BQ4AjnPO/RI6pB3QsczxS4ATgb8An+GTkUudc+VnkMTW5s1+f5DjjoNu3fxe6EosRERE\ngFoY0OmcmwBMiFB3SZiy9/BTWOPTF1/AJZf451Gj/G2QBlpFXUREpJR+K1bW5s1+m/SePWHTJvjw\nQ7jpJiUWIiIi5dT5qahRkZ3t54rm5fl9Qm6/HZo2jXVUIiIicUl/dlfk119hwADo1cv3UHzyid/d\nVImFiIhIREouwikuhokToXNn+Pe/Ydw4n1h06xbryEREROKekouynPN7gfToAVdeCaeeCt98A9dc\nA410B0lERKQylFyU+uADv9nYiSfCLrvA3Lnw2GOwxx6xjkxERCShKLn48EM46SQ44ghYt873XMya\nBQcfHOvIREREElL9TC6c8ytq9u4Nhx0G330HTz0FOTnQt6+W7xYREamB+pVcFBbCtGl+rYrjj4eN\nG2H6dMjNhXPP1ZoVIiIiUVA/RikuWeJnfzz6KKxcCcceC2++CX36qJdCREQkyup+cnH++ZCVBTvv\n7JftvvJKP8VUREREAlH3k4tu3eDII+G886BFi1hHIyIiUufV/eTipptiHYGIiEi9ohGMIiIiElVK\nLkRERCSqlFyIiIhIVCm5EBERkahSciEiIiJRpeRCREREokrJhYiIiESVkgsRERGJKiUXIiIiElVK\nLkRERCSqlFyIiIhIVCm5EBERkahSciEiIiJRpeRCREREokrJhYiIiESVkgsRERGJKiUXIiIiElVK\nLkRERCSqlFyIiIhIVCm5EBERkahSciEiIiJRpeRCREREokrJhYiIiESVkgsRERGJKiUXIiIiElVK\nLkRERCSqlFyIiIhIVCm5SCBZWVmxDqHW1Je2qp11i9pZt9SXdgYhsOTCzFqb2VNmttbMVpvZI2bW\nooLjG5nZCDP7wsx+M7OfzGyqmSUHFWOiqU8/6PWlrWpn3aJ21i31pZ1BCLLn4mkgBegDnAgcCUyq\n4PjmQDdgKNAd+CvQBXgpwBhFREQkyhoF8aZm9kfgOCDdOfdpqOwa4FUzu8k5t6z8Oc65daFzyr7P\n1cBHZtbBObc0iFhFREQkuoLquTgUWF2aWIS8BTjg4Cq8zy6hc9ZEMTYREREJUCA9F0A7YEXZAudc\nsZmtCtXtkJk1Be4FnnbO/VbBoUkACxYsqGaoiWPt2rXk5OTEOoxaUV/aqnbWLWpn3VIf2lnmd2dS\nNN/XnHOVP9hsOHBLBYc4/DiLvwEXOudSyp2/HPinc66isReYWSNgOpAMHF1RcmFm5wJPVa4FIiIi\nEsZ5zrmno/VmVe25uA+YsoNjFgHLgD3KFppZQ2DXUF1EocTiOaAj8Ocd9FoAzATOA5YAhTs4VkRE\nRLZKAvbG/y6Nmir1XFT6Tf2AzlygZ5kBnccCrwEdwg3oDB1Tmlh0wvdYrIp6cCIiIhKoQJILADN7\nDd97cSXQBHgM+Ng5d0GZY74CbnHOvRRKLF7AT0c9iW3HbKxyzhUFEqiIiIhEVVADOgHOBR7EzxIp\nAZ4Hrit3zH5Aq9DrPfFJBcBnoWfDj+M4GngvwFhFREQkSgLruRAREZH6SXuLiIiISFQpuRAREZGo\nSojkwswGmtliMysws7lm1quCY/9qZm+Y2YrQpmlzQjNV4l5V2lnuvMPNrMjMEmK1l6q208yamNk9\nZrbEzArNbJGZXVxL4VZbNdp5npl9ZmYbzOxnM3vUzHatrXirw8z+ZGYzQhsNlpjZKZU4p7eZZYe+\nl9+Y2UW1EWtNVLWdCX4dqvL3tMy5CXMtqubPbsJdi6rZzhpfi+I+uTCzs4HRwGD8hmafAzPNrE2E\nU44E3gD6Aj2Ad4GXzezAWgi32qrRztLzWgFT8QNn41412/kcflDvJUBnIAP4OuBQa6Sq7TSzw/Hf\nx38BqcAZwEHA5FoJuPpa4AdgX4UffF0hM9sbeAV4GzgQGAc8YmbHBBdiVFSpnSTodSikqm0FEu9a\nRPXamXDXIqr+fzQ61yLnXFw/gLnAuDJfG7AUGFSF95gP3BHrtgTRTiALv5PsYCAn1u2IdjuB44FV\nwC6xjj3gdt4ILCxXdjXwQ6zbUoU2lwCn7OCYEcAX5cqygNdiHX802xnhvLi/DtWkrYl2LapqOxP1\nWlSNdkblWhTXPRdm1hhIx/+VA4DzLX0LvzlaZd7DgJ3xPxRxqbrtNLNLgH3w/6HjXjXbeTIwD7jF\nzJaa2ddmNsrMoroOfjRVs50fAh3NrG/oPdoCZwKvBhttrTuE7f+ynUkl/z8nqkS4DtVEol2Lqinh\nrkXVFJVrUZDrXERDG6AhsLxc+XKgSyXf42Z8t9CzUYwr2qrcTjPbDxgGHOGcK/HXrrhXne9nJ+BP\n+KXdTwu9x8P4peQvDSbMGqtyO51zc8zsfOCZ0MWqETAD/xdDXdKO8P8uLc2sqXNuUwxiqg2JcB2q\nlgS9FlVHIl6Lqixa16K47rmoKfObmv0DONM5tzLW8USLmTXAb9Y22Dn3XWlxDEMKUgN8V965zrl5\nzrnXgRuAi8zvnFsnmFkqfvzBEPw9+uPwfwlWuMmfxL+6eh0CXYvQtSiieO+5WAkUA23Llbdlxxug\nnYMfgHKGc+7dYMKLmqq2c2egJ9DNzB4KlTXA975uBo51zs0KKNaaqM73Mx/4yW27gd0C/AWsA/Bd\n2LNiqzrtvBX4wDl3f+jr+WZ2FTDbzG53zpX/az9RLSP8v8u6uthrkWDXoepI1GtRdSTitag6onIt\niuueC+f3E8kG+pSWhe5d9gHmRDrPzDKAR4FzQtllXKtGO9cBXfH7sBwYekwEvgq9/ijgkKulmt/P\nD4D2Zta8TFkX/F8QSwMKtUaq2c7mwJZyZSX40d116S/BDynz7xJybKi8Tkm061A1JeS1qJoS7lpU\nTdG5FsV69GolRreeBWwELgT+iO+a+RXYPVQ/HJha5vhzgc3AAPxfRKWPlrFuSzTbGeb8hBihXY3v\nZwvge+AZIAU/xe9rYGKs2xLldl4EbAr93O4DHA58DMyJdVt20M4W+F8i3UIXoOtDX3eM0M69gfX4\nWSNd8NPjNgN/iXVbotzOhLwOVaetYc5PlGtRVb+niXotqmo7o3ItinnDK/mPcxWwBCjA/4XTs0zd\nFOCdMl+/i++SLv94LNbtiGY7w5ybEP+hq9NO/HzymcBvof/cI4GmsW5HAO0cCHwZaudS/Fzz5Fi3\nYwdtPCp0wQr7/y1CO4/E9+wUAAuBC2Ldjmi3M8GvQ1X+npY7PyGuRdX82U24a1E121nja5E2LhMR\nEZGoiusxFyIiIpJ4lFyIiIhIVCm5EBERkahSciEiIiJRpeRCREREokrJhYiIiESVkgsRERGJKiUX\nIiIiElVKLkRERCSqlFyIiIhIVCm5EBERkaj6f+QM0bPWGrzhAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb39e89d080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sb, lh hemi\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgsAAAFkCAYAAACuFXjcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEpGSBREREYno/wEUko9NNBeyzwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb3a10614e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "az, lh hemi\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgsAAAFkCAYAAACuFXjcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nICIiIg36f4/b7A6WpKSNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb39fe75ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C36, lh hemi\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgsAAAFkCAYAAACuFXjcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nICIiIgH9P8LI5zyw5133AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb39e444c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "# tracers=['C36']\n",
    "hemi_type=['lh']\n",
    "targ='fsaverage5'\n",
    "sm='10'\n",
    "\n",
    "def sigmoid(x, x0, k, a, c):\n",
    "    y = a / (1 + np.exp(-k*(x-x0))) + c\n",
    "    return y\n",
    "\n",
    "def inv_sigmoid(y,x0,k,a,c):\n",
    "    return np.log(a/(y-c)-1)/-k+x0\n",
    "\n",
    "for tracer,nt in zip(tracers,np.arange(0,len(tracers))):\n",
    "    for hemi in hemi_type:\n",
    "        print(tracer+', '+hemi+' hemi')\n",
    "        \n",
    "        # Load data\n",
    "        fmask=opj(surf_data,'mask.'+targ+'.'+hemi)\n",
    "        fname=opj(BPnd_data,'maps',tracer,'mrtm2.nopvc.' + \n",
    "                  targ + '.' + hemi + '.sm' + sm + '.nii.gz')        \n",
    "        data=np.mean(fs.load_surf_data(fname,mask=fmask),axis=1)\n",
    "        \n",
    "        xdata=np.linspace(data.min(),data.max(),10000)\n",
    "        ydata=np.array([np.sum(data<=p)/len(data) for p in xdata])\n",
    "\n",
    "        # Fit sigmoid to data        \n",
    "        p0=[np.ptp(data)/2+data.min(),1/np.ptp(data),1,0]\n",
    "        popt, pcov = curve_fit(sigmoid, xdata, ydata,p0=p0)\n",
    "        yh=sigmoid(xdata,*popt)\n",
    "        ymax=ydata.max()\n",
    "        xh2=inv_sigmoid(ymax/2,*popt)\n",
    "\n",
    "        plt.plot(xdata,ydata,'b')\n",
    "        plt.plot(xdata,yh,'r')\n",
    "        plt.plot([xh2,xh2],[0,ymax],'k--')\n",
    "        plt.plot([popt[0],popt[0]],[0,ymax],'r--')\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Run levelset on mean BPnd maps\n",
    "\n",
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "# tracers=['cumi']\n",
    "hemi_type=['lh']\n",
    "targ='fsaverage5'\n",
    "sm='10'\n",
    "\n",
    "data=np.empty(len(tracers),dtype=object)\n",
    "for tracer,nt in zip(tracers,np.arange(0,len(tracers))):\n",
    "    for hemi in hemi_type:\n",
    "        # Load data\n",
    "        fmask=opj(surf_data,'mask.'+targ+'.'+hemi)\n",
    "        fname=opj(BPnd_data,'maps',tracer,'mrtm2.nopvc.' + \n",
    "                  targ + '.' + hemi + '.sm' + sm + '.nii.gz')        \n",
    "        data=np.mean(fs.load_surf_data(fname,mask=fmask),axis=1)\n",
    "        \n",
    "        # Load neighborhood structure\n",
    "        fneigh=opj(surf_data,'neigh.cortex.'+targ+'.'+hemi+'.list')\n",
    "        neigh,_=fs.load_surf_neighborhood(fneigh)\n",
    "        \n",
    "        # Perform levelset two-sided clustering\n",
    "        clust=fs.surf_levelset_mean(data,neigh)\n",
    "        dest=opj(cluster_data,'levelset','test.mean')\n",
    "        fs.assert_dir(dest)\n",
    "        fname=opj(dest,tracer+'.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(clust,fname,mask=fmask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/data1/vbeliveau/software/anaconda3/lib/python3.5/site-packages/nibabel/spatialimages.py:411: UserWarning: Using large vector Freesurfer hack; header will not be compatible with SPM or FSL\n",
      "  hdr.set_data_shape(shape)\n"
     ]
    }
   ],
   "source": [
    "# Bootstrap levelsets mean borders\n",
    "\n",
    "tracers=['cumi','dasb','sb','az','C36']\n",
    "# tracers=['cumi']\n",
    "hemi_type=['lh']\n",
    "targ='fsaverage6'\n",
    "sm='10'\n",
    "maxiter=100\n",
    "\n",
    "dest=opj(cluster_data,'levelset','test.mean')\n",
    "fs.assert_dir(dest)\n",
    "\n",
    "for tracer,nt in zip(tracers,np.arange(0,len(tracers))):\n",
    "    for hemi in hemi_type:\n",
    "        # Load data\n",
    "        fmask=opj(surf_data,'mask.'+targ+'.'+hemi)\n",
    "        fname=opj(BPnd_data,'maps',tracer,'mrtm2.nopvc.' + \n",
    "                  targ + '.' + hemi + '.sm' + sm + '.nii.gz')        \n",
    "        data=fs.load_surf_data(fname,mask=fmask)\n",
    "        N=data.shape[1]\n",
    "        \n",
    "        # Load neighborhood structure\n",
    "        fneigh=opj(surf_data,'neigh.cortex.'+targ+'.'+hemi+'.list')\n",
    "        neigh,_=fs.load_surf_neighborhood(fneigh)\n",
    "        \n",
    "        clust=np.ndarray(maxiter,dtype=object)\n",
    "        clust_prob=np.zeros(data.shape[0])\n",
    "        for ni in np.arange(0,maxiter):        \n",
    "            mdata=np.mean(data[:,resample(np.arange(0,N))],axis=1)\n",
    "            clust[ni]=fs.surf_levelset_mean(mdata,neigh)\n",
    "            clust_prob+=clust[ni]==-1  \n",
    "        clust_prob=clust_prob/maxiter\n",
    "        \n",
    "        fname=opj(dest,tracer+'.all_iters.'+targ +'.' +hemi+'.sm' + sm) \n",
    "        np.savez(fname,clust)\n",
    "        \n",
    "        fname=opj(dest,tracer+'.all_iters.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(np.vstack(clust).T,fname,mask=fmask)\n",
    "        \n",
    "        fname=opj(dest,tracer+'.prob.'+targ +'.' +hemi+'.sm' + sm + '.nii.gz') \n",
    "        fs.save_surf_data(clust_prob,fname,mask=fmask)"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  },
  "nbpresent": {
   "slides": {},
   "themes": {
    "default": "e62083c0-5297-406b-ac91-6339d2c1cda5",
    "theme": {}
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}