script_validateSegmentation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "vital-broadcast",
   "metadata": {},
   "outputs": [],
   "source": [
    "import SimpleITK as sitk\n",
    "import numpy as np\n",
    "import scipy\n",
    "import os, sys, time, datetime\n",
    "import pathlib\n",
    "import matplotlib.pyplot as plt\n",
    "import re\n",
    "import Metrics\n",
    "import ImageProcessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "blond-citation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0005\n",
      "bone\n",
      "[-178.71, 200.17, -189.453, 189.427, -942.37, -671.37]\n",
      "DSC for bone: [0.9049041333461798, 0.9515235457063712, 0.9597197898423818]\n",
      "Jaccard for bone: [0.8263241245820869, 0.9075297225891678, 0.9225589225589226]\n",
      "MCCD for bone: [1.049453961161892, 0.24053928247005596, 0.9617523666687232]\n",
      "lung\n",
      "[-178.71, 200.17, -189.453, 189.427, -942.37, -671.37]\n",
      "DSC for lung: [0.9920727089118909, 0.9826573469925453, 0.9896433660123769]\n",
      "Jaccard for lung: [0.9842701132151087, 0.9659059748331861, 0.9794990528309829]\n",
      "MCCD for lung: [3.1030607597695785, 5.165562925153485, 3.5128225299024973]\n",
      "0404\n",
      "bone\n",
      "[-213.86700000000002, 227.47699999999998, -220.703, 220.641, -352.431, -32.43099999999998]\n",
      "DSC for bone: [0.8413793103448276, 0.809499747347145, 0.8101535187718498]\n",
      "Jaccard for bone: [0.7261904761904762, 0.6799660441426146, 0.6808891159938681]\n",
      "MCCD for bone: [2.471188044898324, 3.639888182736607, 5.152597822135636]\n",
      "lung\n",
      "[-213.86700000000002, 227.47699999999998, -220.703, 220.641, -352.431, -32.43099999999998]\n",
      "DSC for lung: [0.9944997270856951, 0.9960790720470512, 0.96]\n",
      "Jaccard for lung: [0.9890596291965926, 0.9921887713588283, 0.9230769230769231]\n",
      "MCCD for lung: [1.4946895329787608, 0.27856683205725485, 0.4237467166979369]\n",
      "0612\n",
      "bone\n",
      "[-177.7515625, 182.2484375, -180.3515625, 179.6484375, -307.1015625, 27.8984375]\n",
      "DSC for bone: [0.8976736619489182, 0.6723910171730515, 0.8807947019867549]\n",
      "Jaccard for bone: [0.814344746162928, 0.5064676616915423, 0.7869822485207101]\n",
      "MCCD for bone: [0.9495134359562918, 2.23916131532461, 0.9001621888879836]\n",
      "lung\n",
      "[-177.7515625, 182.2484375, -180.3515625, 179.6484375, -307.1015625, 27.8984375]\n",
      "DSC for lung: [0.9940898650218082, 0.9949303538908303, 0.9788243668070467]\n",
      "Jaccard for lung: [0.9882491789819376, 0.9899118511263467, 0.9585269516728625]\n",
      "MCCD for lung: [2.6608777590021373, 3.4508146357686846, 5.89062496489109]\n",
      "BONE:Mean DSC: 0.8586710473852756, Mean Jaccard: 0.7612503402702574, Mean MCCD (mm): 1.956028511137791\n",
      "LUNG:Mean DSC: 0.9869774229743604, Mean Jaccard: 0.9745209384769742, Mean MCCD (mm): 2.8867518506912697\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD1CAYAAACrz7WZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAA6j0lEQVR4nO3dd3hUVfrA8e87JT1AKIHQW+hNRVCxg4qigrv2XrGAYC+rrrsruir7c3VVRKy42BB17QhiQZCuFAMCoYdekpCeKef3x0wgQEgmybTcvJ/nyZOZO7ecO/POO+eee+85YoxBKaWUtdgiXQCllFLBp8ldKaUsSJO7UkpZkCZ3pZSyIE3uSillQZrclVLKgjS513MicpGIbBGRfBE5JtLlCSUR+YuIvB7gvH8TkSmhLlMgRORtERnnf3yKiKwOZN4abitfRDrWdHkVPTS5W4SIbBSRITVY9F/AaGNMEpAtIkZEHEEuXoVE5HoRmRPE9f0oIsX+BLVHRD4RkbSy140xTxljbg7Stmr6fteKMeZnY0zXYKzL/34d8n4YY5KMMeuDsX4VWZrcVTsgI9KFCKKyH6rOQBK+Hy+l6h1N7hYnIjYReUhE1onIXhGZKiKNRSRWRPIBO7BMRNYBs/2L5fhrvycGsP4BIjJPRHJEZLuIvCQiMf7X2h9+JFBWWxSR7sBE4ET/tnL8rzcUkXdEZLeIbBKRR0Wk2nFqjMkB/gf0K7ftQ5paRORa/zb2ishjFdTGY/xlyRORDBHp71/uv0Bb4At/2R8I4H1aJSLnl3vu8B9dHOt//pGI7BCRXBGZLSI9j7Ke00Ukq9zzY0TkV38ZPwTiyr2WIiJf+t/LbP/j1v7XngROAV7y78NL/ulGRDr7Hx/1syg76hKRf/nXvUFEzq3qfVDho8nd+sYAI4DTgJZANvCyMabEX8MF6GuM6QSc6n/eyH94Pk9E2voTd9ujrN8D3A00BU4EBgN3VFUoY8wq4DZgnn9bjfwvvQg0BDr6y3wtcEN1dhhARJoAfwIyj/J6D2ACcBWQ5t9mq8NmuxD4AGgEfA685C/7NcBm4AJ/2Z/1r3O5iFx5lCK9D1xR7vk5wB5jzK/+598A6UAq8CvwbgD7GIPvB+y/QGPgI+DP5WaxAW/hOzprCxSV24dHgJ/xH+kYY0ZXsImqPouBwGp8n/2zwBsiIlWVW4WHJnfruxV4xBiTZYwpAf4GXBxou7oxZrMxppExZvNRXl9ijJlvjHEbYzYCr+JLBNUmInbgMuBhY0yef33/B1xTjdX8R0RygT34ks6dR5nvYuALY8wcY0wp8Ffg8I6W5hhjvjbGePAl0L6VbdgY08cY895RXn4PuFBEEvzPr/RPK1v2Tf8+l31GfUWkYWXbA04AnMDzxhiXMWYasKjcOvcaYz42xhQaY/KAJwnwswnws9hkjHnN//5Mxvcj2TyQ9avQ0+Rufe2AT/217xxgFb7adlC+hCLSxX+4v0NE9gNP4UuqNdEUiAE2lZu2iSNr1JUZY4xpCPQBUoDWR5mvJbCl7IkxphDYe9g8O8o9LgTianqy2RiTie+9v8Cf4C/En9xFxC4iT/ubzvYDG/2LVfU+tgS2mkN7/zvw3olIgoi86m9S2Y+v2a2RP3FXJZDP4sD743//wHeeQ0UBTe7WtwU411/7LvuLM8ZsrWDemnQR+grwB5BujGkA/AUoOzQv8P9PKDd/i0q2twdw4ftBKtMWqKislTLGrADGAS8fpalgO+USv4jEA02qs4nqlomDTTPDgZX+hA++WvxwYAi+ZpD2ZcWqYn3bgVaH7V/55rN7ga7AQP9nU9bsVjZ/ZfsQtM9CRYYmd2txikhcuT8HvpOWT4pIOwARaSYiw4+y/G7Ai6+NNVDJwH4gX0S6AbeXvWCM2Y0vGVztr53eCHQqt+xOoHXZCVj/4f1Uf3mT/WW+B5jiL3vZCdr2AZZtMr427AsreG0avlr0Sf7t/52qk2l5O6ne+wS+9vuz8b1H5ZtvkoESfEcOCfiOfgIxD3ADY/wnaP8EDDhsvUX4TpA3Bh4PdB+q+ixU9NPkbi1f4/syl/39DXgB38nAGSKSB8zHdyLsCP5D6yeBuf5mnBP8J1TzKzmheh++mmce8Brw4WGv3wLcjy9x9QR+Kffa9/guw9whInv80+7EV+NfD8zBlwTf9L/WBl/TQEC1R39b+n+Axyp4LcO/rQ/w1YDzgF34kmwg/gk86n+f7gPwX1FzVSXl2Y4vIZ/Eoe/TOxzcr5X4PqMq+ffvT8D1+E6UXwZ8Um6W54F4fLXw+cD0w1bxAr7zL9ki8p8KNlHZZ6GinOhgHaquEJFHgd3GmFdDsO4kIAdf89KGYK9fqXDT5K7qLRG5AJiFrznm//Ad0Rxr9EuhLECbZVR9NhzY5v9LBy7XxK6sQmvuSillQVpzV0opC9LkrpRSFqTJXSmlLEiTu1JKWZAmd6WUsiBN7kopZUGa3JVSyoI0uSullAVpcldKKQvS5K6UUhakyV0ppSxIk7tSSlmQJnellLIgTe5KKWVBmtyVUsqCNLkrpZQFaXJXSikL0uSulFIWpMldKaUsSJO7UkpZkCZ3pZSyIE3uSillQZrclVLKgjS5K6WUBWlyV0opC9LkrpRSFqTJXSmlLEiTu1JKWZAmd6WUsiBN7kopZUGa3JVSyoI0uSullAVpcldKKQvS5K6UUhakyV0ppSxIk7tSSlmQJnellLIgTe5KKWVBmtyVUsqCNLkrpZQFaXJXSikL0uSulFIWpMldKaUsSJO7UkpZUMiSu4gMFZHVIpIpIg+FajtKhZPGtaorxBgT/JWK2IE1wFlAFrAIuMIYszLoG1MqTDSuVV0Sqpr7ACDTGLPeGFMKfAAMD9G2lAoXjWtVZ4QqubcCtpR7nuWfplRdpnGt6gxHiNYrFUw7pP1HREYCIwHs2I9LoEGIiqIU5JG9xxjTrJarqTKuQWNbhU8xBZSakoriMmTJPQtoU+55a2Bb+RmMMZOASQANpLEZKINDVBSl4DszbVMQVlNlXIPGtgqfBWbWUV8LVbPMIiBdRDqISAxwOfB5iLalVLhoXKs6IyQ1d2OMW0RGA98CduBNY0xGKLalVLhoXKu6JFTNMhhjvga+DtX6lYoEjWtVV+gdqkopZUGa3JVSyoI0uSullAVpcldKKQvS5K6UUhakyV0ppSxIk7tSSlmQJnellLIgTe5KKWVBmtyVUsqCNLkrpZQFaXJXSikL0uSulFIWpMldKaUsSJO7UkpZkCZ3pZSyIE3uSillQZrclVLKgjS5K6WUBWlyV0opC9LkrpRSFlRlcheRN0Vkl4j8Xm5aYxGZKSJr/f9Tyr32sIhkishqETknVAVXqrY0tpWVBVJzfxsYeti0h4BZxph0YJb/OSLSA7gc6OlfZoKI2INWWqWC6200tpVFVZncjTGzgX2HTR4OTPY/ngyMKDf9A2NMiTFmA5AJDAhOUZUKLo1tZWU1bXNvbozZDuD/n+qf3grYUm6+LP+0I4jISBFZLCKLXZTUsBhKBZ3GtrKEYJ9QlQqmmYpmNMZMMsb0N8b0dxIb5GIoFXQa26pOqWly3ykiaQD+/7v807OANuXmaw1sq3nxlAo7jW1lCTVN7p8D1/kfXwd8Vm765SISKyIdgHRgYe2KqFRYaWwrS3BUNYOIvA+cDjQVkSzgceBpYKqI3ARsBi4BMMZkiMhUYCXgBkYZYzwhKrtStaKxraxMjKmw2TCsGkhjM1AGR7oYysK+M9OWGGP6h3u7GtsqlBaYWew3+yo6H6R3qCqllBVpcldKKQvS5K6UUhakyV0ppSxIk7tSSlmQJnellLKgKq9zV9ZnS0zEfWwXCtNiyelkozC91Dc910Hz+dDo92zMxiy8BQURLqlS1WNLTqbwtO5kd3GQ16v04AseodlcB00X7YMdu/HsPbz/uLpPk3uQ2OLikHatKU1rwJ4+cQemJ23z0CBjH3i8eDdswbhdEMl7C0RAbDjat2HnmWnkdoZzhyzmgdQXSbMnYJdDD+Y8l3rJcJXy5t6T+XzFcTSeE0PqL3vxrtuEKdFOseoDe5PGuLu0YVf/xEN72DGQurgAx9786Ihtmx1xOrB1aseuk5qw9yQXw/suZVzz/5Bkiztids95vtieX9SRl1efhpmdQtqcPGTZGkvEtt7EVFM2O/aUhpT2ac+WM2M59ezl3N9iBo1skGpPPDBbvreY3R43ecbBpN2nMWNdV9iYSOJmocnKYmI37sG7ey/ewsJqfzFsib7tVLSsvXkqpd1bH3jujrez7RQHnU/cRKuEXK5qOo9BcS6c1eyS3GU8bHAX88z2c5g7ozcd39+Dd+0GjNtdrfWEm97EVA02O452rcnr25wt5xnGDprJZcm/k+ZIOmLW7e58dnt9sZ2R04KsJS1J8se2M7sIMjdjSkqqHR9lsQ3gLSoGr+9mYFtCAt5enfAkOA+87o63s+UcO6eekEFjZwG3Nv2ZDo64asc2wAZXPuO2D2XujN60+zIf+XVVVMd2ZTcxaXKvJltCAoWDe7H9yhIePeYrLkjcTIo9odrrcRkPJcbFnOKGrC1pwSurTsGdmUyzXw0Nf9sFu/fiyck9ML84HNibNaWwT2t2DojB1b2Qm3vPJdbmYuLH59Lu8fkHErw9JYWmX3uY0GbGwXJjI8EWU/s34DCzi+HO5VeQPKUBDWaswrN/f9C3EQya3KtmS0ggZ0QfCi/NZXyvaZwcV1CjmCmL7U1uw8+Fnflo23FsWdyKpksNyesLsG/aiWfX7gPxaktOxpaUeERsN7QX4sHG1AfPJWHmcjY+dCyDhy3hgeazaGw72OgQqtje4MrnypXX4ZqWSurnmXh27w76NmpLk3sQ2OLiyPlTP9rcsZbX2n9JQ1t8SLbjMV6Wlrr5vqA7n2X1PTA9NSGPMa1m0Ssmj6bljgwAzlp1AfahOzGu0gNl3TOtLe/2fosOjoOHozWpyQSq0FvKYztP4MtvBtL59a24N24O2bZqQpP70YUrtgHWufJZUdqCFzYOxuXxxeOJqRu4oNFv9HAWHBHbT+3pypxLesPebP54vDN/PetTGtvzOS1+L0lysDvlw5sTg6nQW8r7eW159uOL6PTeXjyr1ka2+akcTe61ZO/Rhc3jnMwd8HpIA7+6Cr2lnP375SQ/FId36cpDXrMlJ2O6tqO4ua+87jgbRddn89Mx71TY/hgsHuPlo/wmPD7tcjq/Fj1JXpN7xcxJfdl2nzvqYttlPNy25TS23t4O81uGb6II9oYNwBlDwcAO7OntwFkApclwyaU/8demK0Ka5D3Gy6ISw7UfjCZ9UnTEtib3GpLYWHbdeCwP3f0elyblVr1AGK0qLeTiV++j7Usr8OblBbSMLTGRvZf0Ie6KHYztMIvzE/cSK86qF6wBj/HyQX4znnvuUpq9ueTAUUWkaHI/lL1BAzbc3YvXr3uJQXHRdUX0qtJCLvzwXtKfXR3wVSyO1q1Yd0s7hp0/n5ubzKF7TPWbSgMVTbGtyb0G7N3T2fKUk3nHvxnSmm515XqLOH7OrbT/jyDzltf48NDRqiXZJ7dlxyDDgOPW8krbL2t07qAqud4i+n05lh5/34x7+46grz9QmtwPsndPx/VSMZ91/TgkbdU1le8t5saNw9jzjw44Zy6pUWzbEhOhQxuyhjam9Ph8bu81m1GN1oWkSTLXW8RJC26m/ei9EYttTe7V5Dq7Pze9+ClXJe+NdFEOMaPQyb2v3kKr5xcHtbZgS0gg94I+JI/MYnynafSJqfzHzGU8ZLmL6OA88uqJozlr1QXEXOvBvTUygxdpcveJ1theXlrMJe/eTccnl/mu/goSR4vmbLq+E+dcMp9HU+eEpAITydjW5F4NrrP7M+qlqfw5KXqu+tjjKWBExjUkPRKPWZIRsu2Iw4F3YC8SntrOR52/PmptZ7M7n/s2D2dqx1nVWv85q87Hca03Il+Cep/cbXayrx3Ao49M5sLE4CXP2ioxLoauvBj+1YyYGTWrrQfClpDA7qv68tzDr3BqCA7EIxXb2p97gFxDjouqxF5iXPx9dw/O/et9JA3fGtLEDmDcbmTuUkovKKLrp3fgMd4K50sWG9kl1a8Bfdv9S9zv2HCktahtUVV12OxkPTiQKX//V1Ql9hmFTvo/P5a44buJ+XZxSK9A8RYW0uS1efxzxOX0nHfVUWO7pgKNbXt6R8QRnntHNbn7uc7uz6gJH0VFYs/1FvH47p4MeHYsCwan0fiteWG9Y86zfz/dn9nKY7v6Vfh6ij2Bmd2/qNG6v+3+JWufa444o6et1+p23TGQH+4YTxdnYtUzh1iJcfFidjs6Tb2N54cNp+X4X4LaDFMV7/I/aH9XLu/mpQZ93YHEthQWYw9T5UaTO9HTFFNiXHSZfS0XjBrLojOb0+KFX/DsiUzbqHtLFr/e2JvphbFVz1xNi06ZyPp/HOfrCkGFlOvs/jw+9p1D7pqOhLLYPvmxMXxzWmc63zUfz+rMiJTFvSWLl/55CctLi4O+7kWnTGTduOOOmuDdW7eBywW20N1zUkaT+wl9oiKxzyqy0/+FsXS6aR3xny2Mio6MzG8ZjPr8hqCvt6EtnmlX/Jv8iwcEfd3qIPdgXzPjiMT8iJZjszufbl/eQaeb1tH4zXkRq7CUlzJ5HiN+uiPo621oi2fulf8i57JjjzqPe8dOX3cKIa7cVJncRaSNiPwgIqtEJENExvqnNxaRmSKy1v8/pdwyD4tIpoisFpFzQrkDtWFLTMQ9LifiiX16YSzPXH01LcfPi7qeF7tO3M3U/IZBX2+fmDiefeYVPGcc/UsQalaObUeL5nT+58qIx3ZGaRFX3nMvXUf9FnWx3e2ZfL4qDP7Z1VR7Iv/3jwmVxrY4HDCwd9C3XV4gNXc3cK8xpjtwAjBKRHoADwGzjDHpwCz/c/yvXQ70BIYCE0RCeN97TYmQ+dc+fNZtakSLUWJcPPDKTci8ZVFzS3N5njXreOrFqygxrqCve1CcjS7PrsTepVPQ1x0gy8b2Hw90YEKruREtRolxMeKDe0j8eGFUdr7lWbmGsZ9dH5J1VxXbxu3GsT0be3rHkGwfAkjuxpjtxphf/Y/zgFVAK2A4MNk/22RghP/xcOADY0yJMWYDkAlE3/H38b346NLnI36D0m1bzqT1m6siWoaqtJyyikd3huYjnNBqPq3+uwNHi+YhWX9lrBrbJef15/M//Tukt+IH4q+7jif9mT+istJSpsvEXSGpvYMvtttM2Ya9c4cKX3dv2gI5gd1dXhPV+vRFpD1wDLAAaG6M2Q6+LwlQdvq5FbCl3GJZ/mlRQ2JjcT+VS7/Y4J8srI7t7nw2PNENT3Z2RMtRFU92Ngv+cTzb3aFpu53QejYrx7WJ6BU0Vort2Hu20zMmsv3ElBgXM14/Kfpje+16HnjjxqBfGlnm1dbzyHyiAbaEii8dDmVPkwEndxFJAj4G7jLGVNaQV9FZgiN+ukVkpIgsFpHFLsLbMb7p24XX098L6zYrcvHKa4mbsSzSxQhI4swM/rX71JCs2yl2Fp3zAoXn9QvJ+qtipdguHtyHD7pEtqkR4Pl9PUh7L7qPSMu0m7aD6UWh64tmxamvs+2WfiFb/9EElNxFxIkv+N81xnzin7xTRNL8r6cBu/zTs4A25RZvDRxx25YxZpIxpr8xpr+T8Nag19wQX61b52uqqtrAnkXNI96hVqC8BQXM+OCEkK2/qT0R++idSJiPpqwW25sukJDcYl9dE+eeEfW19jKetesZ88X1IVt/rDgZcu38QwYgCYdArpYR4A1glTHmuXIvfQ5c5398HfBZuemXi0isiHQA0oGFwSty7That+Lh078M+XZKjIvO34xkg6vipox8bzEtFnhCXo5gav3NPta4QnfFw3td38V9Ys+Qrf9wGtuhUWJcNJsffeeZK5M2x1DoDV1F68HUnyk4K3yxDYHV3AcB1wBnishS/995wNPAWSKyFjjL/xxjTAYwFVgJTAdGGWOiJovtOrstNzTYUvWMQdBknpMHtwyv8LXvixqTtCwynWjV2IYt/HP70JCtPs2RxLqrbeG8uUljOwRWuzw0WVw3au1lGsxez/t5bUO2/lR7IlsvdoX1xr1ArpaZY4wRY0wfY0w//9/Xxpi9xpjBxph0//995ZZ50hjTyRjT1RjzTWh3IXDicNDquvUhHZGoTKw4KWksFHsq7i/9kd9H4N66PeTlCCZvQQEbH+/KuqMcjQTDlDMnHfXqgmDT2A6N7/J7wPrID2RRHZ7du3n244tCdmIV4P1Bk5B+PUK2/sPVqztUPSf25tn2n1Q9Y5Dkd3JzX5vpFb+WnXBg0N+6JG7OKj7MPe6I6S7jwRWESuyAWMOmS7RjseoKd2xX5tWMk/EWFUW6GFFnQKyT1beF73xIvUru6/8cG9bOkxYN+zenxkG2p5A/ZZ7FHk+59uriuvnWF53egxsaLTlk2pUbzmDguNG8tb/NUZYKnFPs3Hn1Z9pzZDWFO7YrU1oQE9XXtlfE0a4Nj1z8UcjvDXjwlK/DFtt1M8PUgKNFc54898OwbGtJSSmjtw6kqT0Rj/Fy/Af3Unx+CQM+vQfwXUXTYk7dfOs3jfC1jZc3f3FXUlaXcnrC2qBs45aGW9g5LDxNM1Zg79yB/xs2JdLFOCh0LRshs+e01lyVvKvqGWspnLFdNzNMDWwf0ZGLkkL/4QH0i3EQby/FY7zYxUbn4zZD86aI5+DJFFv03Y1dJVtcHMP6LT9i+v+GP89jr70VtJqjXWzsO7luXCIaDbaenxbxzsHKeIyXpr+EZlzeUNrbm7Dc0WsXG67zc7RXyGCxJSRwwk2/hWww6MPZxcb4Fr8dCJbp3b5i3bgkbhr8w4HX81vVvbfe1iKVm5vOPmTaqtJCpuYcz+nxwa2utWyRHbZBDeoye0oKQ66ZH+liHGAXG7npkS5F9Xmbh+9ms9u7zMaWGPq297qXYWqg6LSe/KX5d7Vez6XrBzN668BqnTgsu3b2o4GTuLfJ7wemp56/Jew37NRWbv80ujgPHn3keou4esX1nN9gadC3dW27+Uh8ZG+hrwt2XtKNp1osiHQxDtFuQFadi+3E5eHrY8om4TkfYfnkLs4YCkfn0NpR+ztSGzqL+XlrRzJdgf3KLy0poefM2wFfF7fljxzeSn8fz/Hda12mcNpxgpBgO9j/i9cY8gvjcErdu+rHCsQZg3PE7rAdkQbqnfT3KTqrb6SLUS3eE3MjXYSgs3xyLx7Sl096vxWUdb3WZi7LBrxP95jAD6niM2MPdJdb/hra1o4kMq9x1p3RiERo0OXQG1NS7AmsPuUdjovVIfMioXhIX6b1ejvSxThCmiOJrVe46sxQiraEBE5ruy7SxQg6yyf33TcXBqXWHgiP8R7Sc2LvGCdDLlqEzf82v7G/NT8WHXzL3z9rInJs+G5qqA1Hq5b8rYbjpqrQCGdsV9cvp75E0Tn9Il2MwHRpz58bL4p0KYLO0snd3jyVJ/t8VvWMQeQsVxO3i43/tFx04K7BkQ23HXLi8YQ4O3lPFkV9DUccDtaMacuwhPBdkdHKuQ9bo+CPAGUVjtateLbvx5EuxlGl2hPxjN4T9bFtT0nB+cI+BseHr2lxQNwGaBf6nqItndwLj23HaXHhufwRfMm8aTUHIv6053/Ze82Rd3xGk/0X92f25ePDOvjDaXE5lHQO/gj1VpFzUhvOjo+uYesO90mUx7Y4HGQ+2I2POof3iLSz01CaGvobziyd3DcPs9Wo+9M9ngLezWsSghIdKdWeyDlj5mBv0jgs26uu4gsG8NSTk464cSnUYsVJdpe6dcVFOG0/1URFPzKVifbYzrvoOL67cnzYT0jHipOc9NAf0Vg6uTua1qx/i6b2RK5KDt8I7Y81+5WNt3cL2/YC5ejYnhP+sTDo17AHwil2crrVrVvYw8Zmp2mHfVXPFwWiNbYZ0Jt/PP06bSNwzsIpdvYNDP6YxIezbHK3JSZyW685kS5GQGLFycQbJ7D35hPDcudaQGx2Vj+RwjPNl0asCCOHzMLeNDxHUHWJLTGBmztGdvDrQB2I7VuiJ7ZtcXFsuo+wtrMf7r4Tvg15bFs2uWOzkWQvjnQpAnZqHEx+9DlKzzom0kUBoHBEf749+cWIluG2lOXkndI5omVQtXdqHHz913+x/7LjI10UAHZfcwzzTnw1omW4usGakMe2dZN7HdQzJp7kv2QddTDdcLHFxdFk7EY6hWEowso0tMWzv1101PZU7aTaExnxl1nYm0f2JLmjRXOuvuubiA9FGI7Y1uQeZV7v+DGugZFto8wd0Y/Jnf4H+IZMG7+vExevG8Kk3Jb8WGQj31t84MasUMr3FpO0rQ52MagqdGdKBjlndIxoGTbe0Ik7G60HIhvbQAVDqweX9swURrs8BaTY4iq9yiFWbBQ2jyE5jOU6XG4HGyXGyzpXPldlXE+TG/PwZOfwaUwnpPlASto2Zlf/WOaNeY4kW+j65NjnddPw931o5wbRL5DYdoqd3I62iMW2LSGBiy77mSJTyk63O6Kxvd2dT/NFob2U1bLJ3RQVMWXzQEb2/jRiZfixyMZGV1OuSd7B0lI394y5mz29HEy79V8VdmEwt9jLyNfupe3nSyPaJXbbl1ZwxYIx2Es8eLrEs3NYMxpuKMU5exnezA04MjeQVtSXX2+L49QQ9re0srQJUqAj+hzB5WJJXjtoGLkxeKsb22tcBZz76b10fW1txH6sTWkp0188me8LBhGb7aa0g5Odw5oRu99L8qdLDsR248QBrHdDnxBerbistAnOrfsIZc/f1k3ubjdbNjSD3pErw18zh5N8q4fFH+3kjIarSPp9J3FfbOKGLfdwzV++YlQj32DGc4u9XD3jNrpN2E/rFfPwRngUG29eHo7vl7Bm4gDWXPASTrGzxlXATXfdQ2JWIft6JhGb6+WR+25l17E22p6YxfWtf6GFIyeoVyBM3Ho67i11bBDxMPAWFzPj1wHQel7EyvD39ReQMBJKv15EE3v+IbF92r3zGZe6BKfY2e7O58wFt9NqgpPOPy7AE8HYNm43Td6YByKsm9KPzDNeB3y16Asa3U+TFQVkd0skNtfLqHvGsmOgjVlXjA/J5ZLhiG3LJneAzu+62HxefkSuZc0oLSLv8zTiNy/g+03deeHEeTzzUgxN/hxHwynzmbZnKJ/fv501a1vSbcJ+uqxYFPGkfriUZXZ+GBLHwLj9fJ3fk+QxW5jUaSrN7LG4jIfFJQnMKejCiYlrmbL7JH7M6Mqis1+o9l26FfEYL2tmdKKNd2cQ9sR6Ih3bO+a0ol3WYpbmt+XFlr/w4X/z2X9XLxpOmc+K6U3oMn4kjZvmkfRqI9p9uxTjip7BV+zJybAtjqUlJXR2Gr4vasdHj44n2SYk22IOxPaCwk40tgU/RYYrtqs8oSoicSKyUESWiUiGiPzdP72xiMwUkbX+/ynllnlYRDJFZLWInBPKHaiMY8lqnt45JOzb9RgvF791Ly3eWcHmRwcysPUm7GLjH10/x9a8GQAx0xfBkK10uW0h3uV/RN2Yk/uvOIFrR3/Dt7m9GfLXe5g5uCvrfmkH+K5dTrLFcXq8l0eb/sHgeA9vtf2ZDee+HpTEDjCrKJYO74e2ZlOXY9v5+0a+yA9/l9Flsd1gg2HbR51ItJdgFxtjW85EXL6jNs+evXS58VeaXriWuC8XRlViB8ic1J4XLnqLpnYXx719N+8NPoHBX95LsTGHxPaDTdaGpN39D1cJbb8OfRfDgVwtUwKcaYzpC/QDhorICcBDwCxjTDowy/8cEekBXA70BIYCE0Qic5+0t7CQOVvDOxZnRmkR3SePosPLqzFd2jLtxv/jrbY/A3ByXC57Tmt9cOaqEroI9uapmBP74ujYPmw3gYjDQd4ledyVspGxzWbTaG0x7h076fjEbzyx46ywlGHUwqtwb9wS6s3U2dj2ZGfz6pqTw7rNjNIiOn9xGx1fXceuU9z8dvy7jG/xGwD9Yz1sHZJycGZjKo1vcTiwp3cMe2zbGzSgc/M9DEsoprEthma/eXFnbaXrvcvCFtsjfrkds+yPkG+nyuRufMq6A3T6/wwwHJjsnz4ZGOF/PBz4wBhTYozZAGQCA4JZ6Oowv6RUPVMQbXE3Iv317Xj27IXla7nxsXsOdPMbLzHktxZsyVVfL2BvnsraFwdw6U9L+XDqK9wz8ws2f9gDe7NmNS6bvVFDtt13Eq6z+2NLPHoN231yHz477lXyvcWc/fb92OetAGDfpcfwtxYza7z9QG1259PhZcAb2lNvGtvVs8XdiB7P7sK9YyfdH91Ev+dHH4htB3Y8sQQW2107s35KDx769pMDsb3t467YO9euImZLTIQT+uBo0bzCHwtxxpA3pDvvp39MvreYXh+PocGsPxCHgx03HhuW2M71FtHy/ZiQxzYE2Obur50sAToDLxtjFohIc2PMdgBjzHYRKbs7oRVQflDHLP+0iEjYbij0lh4yglAoDY4vZOJbRZjhDfHk5NL0560UmBigmHfzUkk/Zx19L9vHR1NPo+30XGxrt+DJyzuwvKNdGzZf3Jqrr5vJF42n+3tiTGBwvIflJ73NqUNG0eD93TUq264/92DBXc+zz1vKs7vO4Oet3Sha2piEHb4alhGhJAUSdhjaOuKxIXhjfCeibH268chjk8PSgdgneb1wrNwUlqsqNLYDNzi+kNkfr+eXRwcS+9Ui2nzViEZ3FAGxVca2OJzQK53N5zXkrqv/x8gDV/r4YnvFwPcYMOh2UjI31Khs3pP7kf7vVdydOoE/XE0Zv34om9ankrLMl+SNCO6zcihdaSNBYrAhOIoET04utn49eOG+CWGJ7Xdyu5E4JzxXDAWU3I0xHqCfiDQCPhWRXpXMXtHQQkccn4nISGAkQByhu1us8bJs9oXxC+AUO6c2WcusuHQgF/emLTzwxo0cf8d4+sUVc236HgD+PiqDzbfmMyXnOFbmpx1Y/trUbzgjvth/vfChB1azi2NIWZ5d48BInb+P30odDIqL4T8tF0HLRXDYHeGF3lIGPXUXf84cxnud/scrl0zilsSbeWjIF1yYWFjDLVfP66sH0So7Iyzb0tgOnFPsDEhaz4IiX9B416znytfv5sfbqo7tBs5i7kqdRBdnxUeMS0pKabCp5oNUO7OLuKLJPDo5k+jkLGZYr/9BL+DCg/N4jJfei0dXGNuhvKS3vFdXnxy22K7WqWBjTI6I/IivvXGniKT5azZpQFnH6VlAm3KLtQaOODNmjJkETAJoII2j62xiLaxxFfD5g4OJ3eEf2cUYxECKLY5U+6GHim0dSfyl6WpouvqwtRw6n8t4eDmnEx+OG0qDjJqPdO9dvY4H11zMnD6fHHWeLwub0fKLzbinFHLau9ey5LiprP9z+PrhKDEu4r5uELbtldHYrtoaVwHP33ULsd/7Ytu43dg81YntihP7rCI7jz42hgY/1iK2Mzfy5q5TGdT26J0FflyQQod3NuN+MTKxDWD/OXwD0ARytUwzf60GEYkHhgB/AJ8D1/lnuw4oG/Loc+ByEYkVkQ5AOrAwyOWOWitLm5OYcfASJ1tiIudeOq9WfW/3mnMDMwd3pcH7NQ9+8H0Zt26tvG/tx34bjjtrK7RJ4/96TKvV9moi0+Umde6esGxLY7t6QhHb9+84hvGXXUGD92oZ2yUl/JiZXuk8kY7tLHc+aXPyqp4xSAK5WiYN+EFElgOLgJnGmC+Bp4GzRGQtcJb/OcaYDGAqsBKYDozyH/rWCxck7Gfl480Qh++gaPcVffhLs5p3zzoxpxWdHs3HvaP218Ta4uI4u3flh4TNGuZj79GFmAk5R/Tj7jIeXDX8KPd4CtjgqnqYvrf3nYR3/eYabaMGNLarIRSxnXF1Ombx77Uum71pE+459rtK56kstsPh3dxjsK3eFLbtVdksY4xZDhzRD60xZi8w+CjLPAk8WevSBYFxhvdKNbvYWDTkPxw/cSwpi528+eDzNe6BLttTyJvPXkjK2uDciZg1+lgmp42nosNjj/HixsMPvT9i7ufOCoN/VlECez1JNRrIpKk9kWSbi6UlJfSLPfoIS19k9qJdae2/7IGo67Fd2iyROKnoNEBolMX2SVNuJ35BEpPH/LvGse0yHl6aPJxWK3+pfcFsdjbe1pWbG07Hd8FTxSqL7XB4a9UJtMsLT2yDxe9QBcjrlEyjENxlVpE9ngKa2hNpak8k89xJ5A8toaEtPqBl17gKaGaTQ74sT+0eRJOPlte6nxlxOMi+8niev/1VUv03Ge3yFGCDAzcdlRg3Oz2ldHAmHTX4T47LZfL+VKBmo1TFipN+VYyc596cGHU3dEWrXcfEBu2msUA1tSey6pS3yT858NiuyPPZXWj7dmatrxqxJSayeUxfZo18llip/GoXp9gjltgh/LFt+S5/G/yRQ443lN3zHJTnPfjB2cV3yVWgujgTj6gF/e+7E/AW1K7nOFu/Hmz9qAv/Gzf+kH5fUmxxpJT7cibYYuhQRf/t8RLDt7t7UuiNrjsO66uYPFPjZrLasIutVokdYNLXZ+PZWbvB6239epAzrQULRj0X9jF+6wLLJ3fxGApNeA5dyyfHXZ4CTl9xSY3XVegtpdWPtftRsvfsyun/XcSKge8dEfxOsfuvoa/G+sTGqx2mESuWP+CrE1LnZrPBXXdGGyvjMh7S5tbuR6kstuf1/TikXfPWZZZP7t7MjTyzPfxdgHxd0IE4R82T8+ziZBJ/rd2JxVV3NuTBJmtrtY7DpTmSqv2jECiP8ZKw3fIhGTRm7YaIxHZtLSmBpFW1GIBehFX3JQU9tkPNXhy+8yNQD5K7KSlhybt9wje6it85CeuZ1vWDGi+/3xuHKal584e9QQPuPLnyqweizdJSN62n1+JLX8+YkhLmzohgn9Y1tNndGHbvq/Hy9sYpPDMo/Jcy1saSklI6fpQT1m1aPrkDtHzvD57YfWxYt5nmSKrVOI1PrjwPT05OjZc3xpDnqfxw1WO8bHdXfXliuIxZfTneNTW7/by+6jhlF9MLqzhLHWVqG9t4DS5Tt5oGx6y+HLNyXVi3WS+Su2fvPr56/ZSQ194/yEupVrLc5an4ZKnLeHB+2ahWZ9a9eXn8N6PyPq02uwt5dNvQGm8jmHZ5CnC81DTquoeNdp416xgz7cawH5nWhm1mSq1i25OTwxPLzgtiiUIrUrFdL5I7QIs3l3LmistCuo3Lk7MPOXE5vTD2qAm80FvK83tOrPC1H4riaD4zq9blaT3FWemPTQdnEhPb/FTr7QTD8N+vJf7bpZEuRp3U+Zk/uGVzhZflR53ZxdDip5o3yQBgDC3ejouqo87K/CnjmojEdr1J7t7CQpIfjmd+cXAvHdvlKeD57PYVvtbIVoizwr6mfJcePpG6tMLXbv3hetybat+XedzMZVy88tpK56nNrePBMr/YQ6OHY7XWXkOe7GyyHktnaUnNO94Kl+tm3YL399r3ZR43cxl3b7mw6hkjbElJKTH/SolIbNeb5A5glq7k2qmjg3pt8INbh1LoqbjN84Q4e6Xt7hVddZLrLaLNl8E5q25cpeyb1yIo6woVj/FyxYzb8S5bFemi1GnOWb9y0YzReEwkh1avXL63mDZfBS+2f/uxa1DWFUoXzxiFc9avEdl2vUruGEP6Cxt4OadT0Fb5QItvub/JyqCt753cbiTNyQza+loscJHtCU9XvTXxYk5Heozbqnel1pYx9Bi3nRdzOka6JEc1p7ghycuCN25o2hw3+d7ovc7/g7wUur2YF7HYrl/JHXBv38FrU84LWg2ne0xCUJs2npt7Np69tWyTLCf+l9W8kn1E9ylRIddbxOQJ5/l66lO15t6SFdTYDrYHf/8T7k21P5dUJnHFNqbltw3a+oLJZTyMe+uKoDRB1VS9S+4A7d9ax4Sc8I6tGohcbxFtvwjujQ6e/ft5ffbp1VsmTMnhmnUX0eKNyByyWlW0xna+t5i4TxoFdXg5d9ZW/vHdRUFbXzBdtm4obSeEr5OwitTL5O7esZOJU4ZFpF+Oyty55RwSvg/+KC1dJhcGfLIt31vMf/NC306/sMRF8V9a4C2O3sPquihaY/ut3K40+SL4tdjqxHa4LCxxkf9wKzz790e0HPUyuQO0+3gnC0vCeztwZfK9xax+pWetOwqriFn8OxfNGhVQjTzJFsf1DWrXoVNVCr2lXDd5LDJ3aUi3U1+1/SaHLHdRpItxgMd4mfDhMDzZ2UFfd3ViOxyiKbbrbXL3ZG7kvtU179gr2M5cdg2Npy0LzcqNocfftjNuT2XDg4bPCYuvo/3T2hwTKmbVOq7745pIF+OAQcsupcOLIboaSmP7qOptcsfrIXlcMh/kpUS6JHxVGEfjR2PwFobuqhZ31lbm3XwcXxVGtge9KzecQZu7C7U5JoRMSQmx4xoF/Z6OmviqMI5Gj8SGpNZexp21ldl3n8jy0sjGVLTFdv1N7oDMXcqzz1/O5gje6VboLeUvL9+I+S30I6KbRSt4/OkbWFUamUsjX8xuR+41DXFvCN9QY/WVbc5Sbv3PnfUmth0//MoNT9191DvCQ+2rwjiyR6VFVWzX6+QO0GzifC749wMR+xL0/vFWWr66NGzba/LGfK558l7WBTCeaTDNKHQy7aFzcK/fGNbt1lvG0OKFeVzw7wcidpv+mcuvDF9sG0Ozt5Yw8Ou7wt7PzoxCJ8/efS3epcG73yUY6n1y930JFnDuxAfCHhS3bBlEt8f2hbQ55gjG0OT1+Vw48YGw1eCf2ZvOk3ffQNwXC8OyPeVnDGkvLmTQD2PCfvXMmG3H03isN6yxbVyldLs7g+6fjg5b5eV/BUlRG9ua3AG8Htq99Du9frolbF+CmzafzNarm0fmMM4YWj89j2vG3RvyBH/lhjOYfWGPqAz++sC43XS7M5MeP90Uttj+scjG8sf64Vm7PizbK89bWEj6nQu57In7wxLbk0YMi9rYDji5i4hdRH4TkS/9zxuLyEwRWev/n1Ju3odFJFNEVotInRgqxrN/P+m3rQ/Ll+CWLYPYfk1qRIL/AGNo8sZ8br99LKO3Dgz6Ucs6Vz4dPhtJ7mUJUdUOeTirxzWEP7afvO5aYr9ZFNLtVMof2zc8eg/37zgm6JdJlo9tT8bqoK47mKpTcx8LlL+e6SFgljEmHZjlf46I9AAuB3oCQ4EJIlHQ9WAADnwJ/js6JG3wezwFdPnpOrZeFeHEXsYYYr9ZxPqzEzj2pbF8XlDzwUXK5HqLuHzDmVx37710GbWkLnQtYPm4hoOx3f2nm8j1Bv8a+PKxbZuzNOjrrzZjaDhlPhlnN6bbu6OCEtsu4+HKDWfUmdgOKLmLSGtgGPB6ucnDgcn+x5OBEeWmf2CMKTHGbAAygcpHjYginv376fjwfEb8836m5jcMyjpdxsMze9MZ+rf76Hj9H3gyo2u0IU92Nq3/+QsTzz2H3s/dwf07jqHQW70uSsuS+tAH7ib3rGISpy0I6q3moVCf4hr8Cf7mNZzy3L0sLAnOkZrHeKM7tvfspeMD85h47jmkT7mdt/en1ii2b9h8Cv1evpOcYd46EdsAYgLosUxEpgH/BJKB+4wx54tIjjGmUbl5so0xKSLyEjDfGDPFP/0N4BtjzFEHPWwgjc1Aib7BBmx9urHjCfj22DdItSfWaB3P7E1nyuSzaDN5LZ7du4NcwtCwp6SQe1ZXtp3p5YL+v5Eev4vLGqw8pCZQaAwf7u/Lsv1t+GVRN9rO8BD/Q0Z4Tw5Xw3dm2hJjTP/y00Id1xClsS2C99R+uB/dx4fd3qtxbP9YZOPGr2+h27gNeHaG9q7mYLE3TyX3tI7suKCE246ZzRUNlhEnR96p/nVBO9YUp/Hu/BNp94WJ2theYGax3+yr8Fb7KgciFJHzgV3GmCUicnoA26toQ0f8gojISGAkQBy1P2QKBe/yP2h+aSwXD70b++idvN7lXTo5k6peEH/gT7+F7v/MomXWL0T/7/xBnuxskqbOp8tUWBMbS2Z8e75KPxVjP/jRisuD/LERU1pIums+ANFxA3hgQhXX/nVHd2wbg+2n34idX7vY7jYhh/SVC/HUoe6aPTt3kTR1F50/Er5PaMGs7idiHIc1YHgN9g078O7fT5cS38nSuhTbZQIZZXYQcKGInAfEAQ1EZAqwU0TSjDHbRSQNKPvpzgLalFu+NbDt8JUaYyYBk8BXu6nFPoSUKSkh/rOFyPRYRvW5nU3nJXPcOSu5uNlijo3dQZo9HoDN7iJWlLbgne0nsvqbdNp9vpf0jAW4I1z+2jIlJXhKSmDRikOnc5TMVneEJK6h7sf2FanzOS0uh1hxAuDFy7ziWL7L68mnH55yILbrUoXlCMb4+nFaXHHPjXV63/wCapY5MLOvhlN2+Doe2GuMeVpEHgIaG2MeEJGewHv42iNb4jsplW7M0U/TR+WhayVscXFIYgLubm0pSYkBIG5PMfa1WXjzCzBR1kudqrhZpkyo4hrqaGw3bEDRMe3wOnwHK2IMiRk78e7NxpuXF+ESqvJq1SxTiaeBqSJyE7AZuATAGJMhIlOBlYAbGFXVF6Cu8RYXQ3ExMncf5XtqsdRO1l/1Nq7hYGzHTD+0Db2uH4HWR9WquYdKXavdqLqnspp7KGlsq1CqrOaud6gqpZQFaXJXSikL0uSulFIWpMldKaUsSJO7UkpZkCZ3pZSyIE3uSillQZrclVLKgjS5K6WUBWlyV0opC9LkrpRSFqTJXSmlLEiTu1JKWZAmd6WUsiBN7kopZUGa3JVSyoI0uSullAVpcldKKQvS5K6UUhakyV0ppSwooOQuIhtFZIWILBWRxf5pjUVkpois9f9PKTf/wyKSKSKrReScUBVeqdrS2FZWVZ2a+xnGmH7lRpB/CJhljEkHZvmfIyI9gMuBnsBQYIKI2INYZqWCTWNbWU5tmmWGA5P9jycDI8pN/8AYU2KM2QBkAgNqsR2lwk1jW9V5gSZ3A8wQkSUiMtI/rbkxZjuA/3+qf3orYEu5ZbP805SKRhrbypIcAc43yBizTURSgZki8kcl80oF08wRM/m+SCMB4kgIsBhKBZ3GtrKkgGruxpht/v+7gE/xHYruFJE0AP//Xf7Zs4A25RZvDWyrYJ2TjDH9jTH9ncTWfA+UqgWNbWVVVSZ3EUkUkeSyx8DZwO/A58B1/tmuAz7zP/4cuFxEYkWkA5AOLAx2wZWqLY1tZWWBNMs0Bz4VkbL53zPGTBeRRcBUEbkJ2AxcAmCMyRCRqcBKwA2MMsZ4QlJ6pWpHY1tZlhhzRJNh+AshkgesjnQ5wqgpsCfShQiTaNnXdsaYZuHeqMa2pUXDvh41rgM9oRpqq8tdY2x5IrK4vuxvfdrXo9DYtqho31ftfkAppSxIk7tSSllQtCT3SZEuQJjVp/2tT/takfq2//Vpf6N6X6PihKpSSqngipaau1JKqSCKeHIXkaH+7lMzReShSJentkSkjYj8ICKrRCRDRMb6p1u2G1kRsYvIbyLypf+5Zfe1OjS26/7nXadj2xgTsT/ADqwDOgIxwDKgRyTLFIR9SgOO9T9OBtYAPYBngYf80x8CnvE/7uHf71igg//9sEd6P6q5z/cA7wFf+p9bdl+r8Z5obFvg867LsR3pmvsAINMYs94YUwp8gK9b1TrLGLPdGPOr/3EesApfz4GW7EZWRFoDw4DXy0225L5Wk8Z2Hf+863psRzq5W7oLVRFpDxwDLMC63cg+DzwAeMtNs+q+Voel91VjO/r3NdLJPaAuVOsiEUkCPgbuMsbsr2zWCqbVifdARM4HdhljlgS6SAXT6sS+1oBl91Vju+JFKpgW0X2NdPcDAXWhWteIiBNf8L9rjPnEP3mniKQZY7bXpBvZKDUIuFBEzgPigAYiMgVr7mt1WXJfNbbr0L5G+GSFA1iP7wRE2UmnnpEsUxD2SYB3gOcPmz6eQ0/EPOt/3JNDT8Ssp46ddPLvx+kcPOlk6X0N8P3Q2LbI511XYzsa3rjz8J11Xwc8EunyBGF/TsZ3OLYcWOr/Ow9ogm+w5bX+/43LLfOIf/9XA+dGeh9quN/lvwCW3tdqvCca2xb4vOtqbOsdqkopZUGRPqGqlFIqBDS5K6WUBWlyV0opC9LkrpRSFqTJXSmlLEiTu1JKWZAmd6WUsiBN7kopZUH/D5V4Rhw37f4EAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "savefig_directory = 'C:/Users/nfdlam/OneDrive - The Hong Kong Polytechnic University/images/CXR-GAN/DRR'\n",
    "\n",
    "DSC_bone_list = []\n",
    "Jaccard_bone_list = []\n",
    "DSC_lung_list = []\n",
    "Jaccard_lung_list = []\n",
    "MCCD_lung_list = []\n",
    "MCCD_bone_list = []\n",
    "\n",
    "for patient_code in ['0005','0404','0612']:\n",
    "    print(patient_code)\n",
    "    # Information for each patient\n",
    "    if patient_code == '0005':\n",
    "        axial_slice = 115\n",
    "        coronal_slice = 257\n",
    "        sagittal_slice = 329\n",
    "\n",
    "        origin = [-178.34, -189.083, -942.0] # R, A, I\n",
    "        VoxelSize = [0.74, 0.74, 1.0]  # LR, AP, IS\n",
    "        \n",
    "        validation_bone_file = '0005-bone.seg.nrrd'\n",
    "        validation_lung_file = '0005-lung.seg.nrrd'\n",
    "        auto_bone_file = '0005_bone_autosegmentation.npy'\n",
    "        auto_lung_file = '0005_lung_autosegmentation.npy'\n",
    "        \n",
    "    elif patient_code =='0404':\n",
    "        axial_slice=24\n",
    "        coronal_slice=256\n",
    "        sagittal_slice=272\n",
    "\n",
    "        origin = [-213.436, -220.272, -352.0] # R, A, I\n",
    "        VoxelSize = [0.862, 0.862, 5.0]  # LR, AP, IS\n",
    "\n",
    "        validation_bone_file = '0404-bone.seg.nrrd'\n",
    "        auto_bone_file = '0404_bone_autosegmentation.npy'\n",
    "        validation_lung_file = '0404-lung.seg.nrrd'\n",
    "        auto_lung_file = '0404_lung_autosegmentation.npy'\n",
    "        \n",
    "    elif patient_code=='0612':\n",
    "        axial_slice=31\n",
    "        coronal_slice=239\n",
    "        sagittal_slice=203\n",
    "\n",
    "        origin = [-177.4, -180.0, -306.75] # R, A, I\n",
    "        VoxelSize = [0.703125, 0.703125, 5.0]  # LR, AP, IS\n",
    "\n",
    "        validation_bone_file = '0612-bone.seg.nrrd'\n",
    "        auto_bone_file = '0612_bone_autosegmentation.npy'\n",
    "        validation_lung_file = '0612-lung.seg.nrrd'\n",
    "        auto_lung_file = '0612_lung_autosegmentation.npy'\n",
    "\n",
    "    tissue_range = [\"bone\",\"lung\"]\n",
    "    # VALIDATION FILE PATH\n",
    "    directoryPath_validation = pathlib.Path(\"D:/data/CT-Covid-19-August2020/validation_segmentations/\")\n",
    "    #print(directoryPath_validation)\n",
    "    directoryPath_auto = pathlib.Path(\"D:/data/CT-Covid-19-August2020/auto_segmentation/\")\n",
    "    #print(directoryPath_auto)\n",
    "\n",
    "    for tissue in tissue_range:\n",
    "        print(tissue)\n",
    "        if tissue ==\"bone\":\n",
    "            validation_file = validation_bone_file\n",
    "            auto_file = auto_bone_file\n",
    "        if tissue ==\"lung\":\n",
    "            validation_file = validation_lung_file\n",
    "            auto_file = auto_lung_file\n",
    "\n",
    "        # LOAD VALIDATION FILE\n",
    "        validation_file_path = os.path.join(directoryPath_validation, validation_file)\n",
    "\n",
    "        # Load image\n",
    "        reader = sitk.ImageFileReader()\n",
    "        reader.SetFileName(validation_file_path)\n",
    "        reader.LoadPrivateTagsOn()\n",
    "        reader.ReadImageInformation()\n",
    "        # Read Metadata\n",
    "        for k in reader.GetMetaDataKeys():\n",
    "            v = reader.GetMetaData(k)\n",
    "            #print(f\"({k}) = = \\\"{v}\\\"\")\n",
    "        image = reader.Execute();\n",
    "        validation_array = sitk.GetArrayFromImage(image)\n",
    "\n",
    "        axial_validation = validation_array[axial_slice,:,:]\n",
    "        coronal_validation = validation_array[:,coronal_slice,:]\n",
    "        sagittal_validation = validation_array[:,:,sagittal_slice]\n",
    "\n",
    "        #plt.imshow(axial_validation)\n",
    "        #plt.plot\n",
    "\n",
    "        # LOAD AUTO FILE\n",
    "        auto_array = np.load(os.path.join(directoryPath_auto, auto_file))\n",
    "        axial_auto = auto_array[axial_slice,:,:]\n",
    "        coronal_auto = auto_array[:,coronal_slice,:]\n",
    "        sagittal_auto = auto_array[:,:,sagittal_slice]\n",
    "        \n",
    "        # Extents (required for MCCD)\n",
    "        extents =[origin[0]-0.5*VoxelSize[0], origin[0]-0.5*VoxelSize[0]+VoxelSize[0]*(auto_array.shape[1]),\n",
    "                  origin[1]-0.5*VoxelSize[0], origin[1]-0.5*VoxelSize[0]+VoxelSize[1]*(auto_array.shape[2]),\n",
    "                  origin[2]-0.5*VoxelSize[0], origin[2]-0.5*VoxelSize[0]+VoxelSize[2]*(auto_array.shape[0])]\n",
    "        print(extents)\n",
    "        extents_coronal = [[extents[4], extents[5]], [extents[0], extents[1]]]\n",
    "        extents_axial = [[extents[0], extents[1]], [extents[2], extents[3]]]\n",
    "        extents_sagittal = [[extents[4], extents[5]], [extents[2], extents[3]]]\n",
    "        \n",
    "        # CALCULATE METRICS\n",
    "        DSC_axial = Metrics.DiceSimilarityCoefficient(axial_auto, axial_validation)\n",
    "        DSC_coronal = Metrics.DiceSimilarityCoefficient(coronal_auto, coronal_validation)\n",
    "        DSC_sagittal = Metrics.DiceSimilarityCoefficient(sagittal_auto, sagittal_validation)\n",
    "        Jaccard_axial = Metrics.JaccardIndex(axial_auto, axial_validation)\n",
    "        Jaccard_coronal = Metrics.JaccardIndex(coronal_auto, coronal_validation)\n",
    "        Jaccard_sagittal = Metrics.JaccardIndex(sagittal_auto, sagittal_validation)\n",
    "        MCCD_axial = Metrics.MCCD(axial_auto, axial_validation, extents_axial)\n",
    "        MCCD_coronal = Metrics.MCCD(coronal_auto, coronal_validation, extents_coronal)\n",
    "        MCCD_sagittal = Metrics.MCCD(sagittal_auto, sagittal_validation, extents_sagittal)\n",
    "        \n",
    "        print(\"DSC for \"+tissue+\": \"+str([DSC_axial, DSC_coronal, DSC_sagittal]))\n",
    "        print(\"Jaccard for \"+tissue+\": \"+str([Jaccard_axial, Jaccard_coronal, Jaccard_sagittal]))\n",
    "        print(\"MCCD for \"+tissue+\": \"+str([MCCD_axial, MCCD_coronal, MCCD_sagittal]))\n",
    "        if tissue == \"lung\":\n",
    "            DSC_lung_list.append(DSC_axial)\n",
    "            DSC_lung_list.append(DSC_coronal)\n",
    "            DSC_lung_list.append(DSC_sagittal)\n",
    "            Jaccard_lung_list.append(Jaccard_axial)\n",
    "            Jaccard_lung_list.append(Jaccard_coronal)\n",
    "            Jaccard_lung_list.append(Jaccard_sagittal)\n",
    "            MCCD_lung_list.append(MCCD_axial)\n",
    "            MCCD_lung_list.append(MCCD_coronal)\n",
    "            MCCD_lung_list.append(MCCD_sagittal)\n",
    "        elif tissue == \"bone\":\n",
    "            DSC_bone_list.append(DSC_axial)\n",
    "            DSC_bone_list.append(DSC_coronal)\n",
    "            DSC_bone_list.append(DSC_sagittal)\n",
    "            Jaccard_bone_list.append(Jaccard_axial)\n",
    "            Jaccard_bone_list.append(Jaccard_coronal)\n",
    "            Jaccard_bone_list.append(Jaccard_sagittal)\n",
    "            MCCD_bone_list.append(MCCD_axial)\n",
    "            MCCD_bone_list.append(MCCD_coronal)\n",
    "            MCCD_bone_list.append(MCCD_sagittal)\n",
    "        \n",
    "        \n",
    "        fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "        fig.suptitle('Left: auto, Right: validation')\n",
    "        ax1.imshow(axial_auto)\n",
    "        ax2.imshow(axial_validation)\n",
    "        fig.savefig(os.path.join(savefig_directory,patient_code+\"_\"+tissue+'_autoVSvalidation.png'), dpi=150)\n",
    "\n",
    "# Total Output\n",
    "meanDSC_bone = np.mean(DSC_bone_list)\n",
    "meanJaccard_bone = np.mean(Jaccard_bone_list)\n",
    "meanMCCD_bone = np.mean(MCCD_bone_list)\n",
    "print(\"BONE:Mean DSC: \" + str(meanDSC_bone) + \", Mean Jaccard: \" + str(meanJaccard_bone) + \", Mean MCCD (mm): \" + str(meanMCCD_bone))\n",
    "meanDSC_lung = np.mean(DSC_lung_list)\n",
    "meanJaccard_lung = np.mean(Jaccard_lung_list)\n",
    "meanMCCD_lung = np.mean(MCCD_lung_list)\n",
    "print(\"LUNG:Mean DSC: \" + str(meanDSC_lung) + \", Mean Jaccard: \" + str(meanJaccard_lung) + \", Mean MCCD (mm): \" + str(meanMCCD_lung))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "opening-fundamental",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "widespread-canadian",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(67, 512, 512)\n",
      "57791\n",
      "2.6608777590021373\n"
     ]
    }
   ],
   "source": [
    "print(validation_array.shape)\n",
    "\n",
    "points_auto = ImageProcessing.mask2points(axial_auto, extents_axial)\n",
    "print(points_auto.shape[0])\n",
    "c = (1,0)\n",
    "d = np.linalg.norm(c-points_auto, axis=1)\n",
    "f = Metrics.MCCD(axial_auto, axial_validation, extents_axial)\n",
    "print(f)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}