6.0 - Make_patient_to_single_rows.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_csv('./datasets/ihm/train.csv')\n",
    "test = pd.read_csv('./datasets/ihm/test.csv')\n",
    "val = pd.read_csv('./datasets/ihm/val.csv')\n",
    "train = train.drop('Unnamed: 0', axis=1)\n",
    "test = test.drop('Unnamed: 0', axis=1)\n",
    "val = val.drop('Unnamed: 0', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def split_df(df):\n",
    "    split_set = []\n",
    "    for episode, df_episode in df.groupby('Episode'):\n",
    "        split_set.append(df_episode)\n",
    "    return split_set\n",
    "\n",
    "train_split = split_df(train)\n",
    "val_split = split_df(val)\n",
    "test_split = split_df(test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7eff85fa04a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_episode_lengths(df):\n",
    "    patient_lengths = []\n",
    "    for patient in df:\n",
    "        patient_lengths.append(len(patient))\n",
    "    plt.hist(patient_lengths,200)\n",
    "    plt.axis([0,200,0,10000])\n",
    "    plt.show()\n",
    "    \n",
    "    \n",
    "plot_episode_lengths(train_split)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "500/14681\n",
      "1000/14681\n",
      "1500/14681\n",
      "2000/14681\n",
      "2500/14681\n",
      "3000/14681\n",
      "3500/14681\n",
      "4000/14681\n",
      "4500/14681\n",
      "5000/14681\n",
      "5500/14681\n",
      "6000/14681\n",
      "6500/14681\n",
      "7000/14681\n",
      "7500/14681\n",
      "8000/14681\n",
      "8500/14681\n",
      "9000/14681\n",
      "9500/14681\n",
      "10000/14681\n",
      "10500/14681\n",
      "11000/14681\n",
      "11500/14681\n",
      "12000/14681\n",
      "12500/14681\n",
      "13000/14681\n",
      "13500/14681\n",
      "14000/14681\n",
      "14500/14681\n",
      "500/3222\n",
      "1000/3222\n",
      "1500/3222\n",
      "2000/3222\n",
      "2500/3222\n",
      "3000/3222\n",
      "500/3236\n",
      "1000/3236\n",
      "1500/3236\n",
      "2000/3236\n",
      "2500/3236\n",
      "3000/3236\n"
     ]
    }
   ],
   "source": [
    "def fix_episode_length(df, rows, test=False):\n",
    "    counter = 0\n",
    "    counter_length = len(df)\n",
    "    new_data  = []\n",
    "    for patient in df:\n",
    "        length = len(patient)\n",
    "        patient = patient.copy()\n",
    "        patient['Length'] = length\n",
    "        if length < 25 and test == False : \n",
    "            continue\n",
    "        elif length < rows:\n",
    "            duplicate_data = []\n",
    "            while length < rows:\n",
    "                duplicate_data.append(patient.iloc[0])\n",
    "                length=length+1\n",
    "            new_data.append(pd.concat([pd.DataFrame(duplicate_data), patient], ignore_index=True))\n",
    "        elif length > rows:\n",
    "            new_data.append(patient.iloc[length-rows:])\n",
    "            length-rows\n",
    "        else:\n",
    "            new_data.append(patient)\n",
    "        counter = counter+1\n",
    "        if counter %500 == 0:\n",
    "            print(str(counter) + \"/\" + str(counter_length))\n",
    "\n",
    "    return new_data\n",
    "\n",
    "train_split = fix_episode_length(train_split, 90)\n",
    "val_split = fix_episode_length(val_split, 90)\n",
    "test_split = fix_episode_length(test_split, 90, test=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Hours</th>\n",
       "      <th>Capillary refill rate</th>\n",
       "      <th>Diastolic blood pressure</th>\n",
       "      <th>Fraction inspired oxygen</th>\n",
       "      <th>Glascow coma scale eye opening</th>\n",
       "      <th>Glascow coma scale motor response</th>\n",
       "      <th>Glascow coma scale total</th>\n",
       "      <th>Glascow coma scale verbal response</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>Heart Rate</th>\n",
       "      <th>...</th>\n",
       "      <th>Mean blood pressure</th>\n",
       "      <th>Oxygen saturation</th>\n",
       "      <th>Respiratory rate</th>\n",
       "      <th>Systolic blood pressure</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Weight</th>\n",
       "      <th>pH</th>\n",
       "      <th>Mortality</th>\n",
       "      <th>Episode</th>\n",
       "      <th>Length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.21</td>\n",
       "      <td>Spontaneously</td>\n",
       "      <td>Obeys Commands</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Oriented</td>\n",
       "      <td>304.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>...</td>\n",
       "      <td>77.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>36.6</td>\n",
       "      <td>81.0</td>\n",
       "      <td>7.4</td>\n",
       "      <td>0</td>\n",
       "      <td>10004_episode1_timeseries.csv</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.21</td>\n",
       "      <td>Spontaneously</td>\n",
       "      <td>Obeys Commands</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Oriented</td>\n",
       "      <td>304.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>...</td>\n",
       "      <td>77.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>36.6</td>\n",
       "      <td>81.0</td>\n",
       "      <td>7.4</td>\n",
       "      <td>0</td>\n",
       "      <td>10004_episode1_timeseries.csv</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.21</td>\n",
       "      <td>Spontaneously</td>\n",
       "      <td>Obeys Commands</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Oriented</td>\n",
       "      <td>304.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>...</td>\n",
       "      <td>77.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>36.6</td>\n",
       "      <td>81.0</td>\n",
       "      <td>7.4</td>\n",
       "      <td>0</td>\n",
       "      <td>10004_episode1_timeseries.csv</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.21</td>\n",
       "      <td>Spontaneously</td>\n",
       "      <td>Obeys Commands</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Oriented</td>\n",
       "      <td>304.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>...</td>\n",
       "      <td>77.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>36.6</td>\n",
       "      <td>81.0</td>\n",
       "      <td>7.4</td>\n",
       "      <td>0</td>\n",
       "      <td>10004_episode1_timeseries.csv</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.015</td>\n",
       "      <td>0.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0.21</td>\n",
       "      <td>Spontaneously</td>\n",
       "      <td>Obeys Commands</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Oriented</td>\n",
       "      <td>304.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>...</td>\n",
       "      <td>77.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>36.6</td>\n",
       "      <td>81.0</td>\n",
       "      <td>7.4</td>\n",
       "      <td>0</td>\n",
       "      <td>10004_episode1_timeseries.csv</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Hours  Capillary refill rate  Diastolic blood pressure  \\\n",
       "0  0.015                    0.0                      59.0   \n",
       "1  0.015                    0.0                      59.0   \n",
       "2  0.015                    0.0                      59.0   \n",
       "3  0.015                    0.0                      59.0   \n",
       "4  0.015                    0.0                      59.0   \n",
       "\n",
       "   Fraction inspired oxygen Glascow coma scale eye opening  \\\n",
       "0                      0.21                  Spontaneously   \n",
       "1                      0.21                  Spontaneously   \n",
       "2                      0.21                  Spontaneously   \n",
       "3                      0.21                  Spontaneously   \n",
       "4                      0.21                  Spontaneously   \n",
       "\n",
       "  Glascow coma scale motor response  Glascow coma scale total  \\\n",
       "0                    Obeys Commands                      15.0   \n",
       "1                    Obeys Commands                      15.0   \n",
       "2                    Obeys Commands                      15.0   \n",
       "3                    Obeys Commands                      15.0   \n",
       "4                    Obeys Commands                      15.0   \n",
       "\n",
       "  Glascow coma scale verbal response  Glucose  Heart Rate  ...    \\\n",
       "0                           Oriented    304.0        86.0  ...     \n",
       "1                           Oriented    304.0        86.0  ...     \n",
       "2                           Oriented    304.0        86.0  ...     \n",
       "3                           Oriented    304.0        86.0  ...     \n",
       "4                           Oriented    304.0        86.0  ...     \n",
       "\n",
       "   Mean blood pressure  Oxygen saturation  Respiratory rate  \\\n",
       "0                 77.0               98.0              19.0   \n",
       "1                 77.0               98.0              19.0   \n",
       "2                 77.0               98.0              19.0   \n",
       "3                 77.0               98.0              19.0   \n",
       "4                 77.0               98.0              19.0   \n",
       "\n",
       "   Systolic blood pressure  Temperature  Weight   pH  Mortality  \\\n",
       "0                    118.0         36.6    81.0  7.4          0   \n",
       "1                    118.0         36.6    81.0  7.4          0   \n",
       "2                    118.0         36.6    81.0  7.4          0   \n",
       "3                    118.0         36.6    81.0  7.4          0   \n",
       "4                    118.0         36.6    81.0  7.4          0   \n",
       "\n",
       "                         Episode Length  \n",
       "0  10004_episode1_timeseries.csv     69  \n",
       "1  10004_episode1_timeseries.csv     69  \n",
       "2  10004_episode1_timeseries.csv     69  \n",
       "3  10004_episode1_timeseries.csv     69  \n",
       "4  10004_episode1_timeseries.csv     69  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_split[0].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "500/14523\n",
      "1000/14523\n",
      "1500/14523\n",
      "2000/14523\n",
      "2500/14523\n",
      "3000/14523\n",
      "3500/14523\n",
      "4000/14523\n",
      "4500/14523\n",
      "5000/14523\n",
      "5500/14523\n",
      "6000/14523\n",
      "6500/14523\n",
      "7000/14523\n",
      "7500/14523\n",
      "8000/14523\n",
      "8500/14523\n",
      "9000/14523\n",
      "9500/14523\n",
      "10000/14523\n",
      "10500/14523\n",
      "11000/14523\n",
      "11500/14523\n",
      "12000/14523\n",
      "12500/14523\n",
      "13000/14523\n",
      "13500/14523\n",
      "14000/14523\n",
      "14500/14523\n",
      "train done\n",
      "500/3188\n",
      "1000/3188\n",
      "1500/3188\n",
      "2000/3188\n",
      "2500/3188\n",
      "3000/3188\n",
      "val done\n",
      "500/3236\n",
      "1000/3236\n",
      "1500/3236\n",
      "2000/3236\n",
      "2500/3236\n",
      "3000/3236\n"
     ]
    }
   ],
   "source": [
    "def make_patients_to_rows(df):\n",
    "    new_dataframe = pd.DataFrame()\n",
    "    counter = 0\n",
    "    counter_length = len(df)\n",
    "    for patient in df: \n",
    "        label = patient.iloc[0][\"Mortality\"]\n",
    "        length = patient.iloc[0][\"Length\"]\n",
    "        patient = patient.drop(['Mortality', 'Episode', 'Length'], axis=1)\n",
    "        patient = patient.reset_index()\n",
    "        patient = patient.drop('index', axis=1)\n",
    "        data_row = pd.DataFrame()\n",
    "        for i in range(0, len(patient)):\n",
    "            \n",
    "            columns = [str(col) + str(i) for col in patient.columns]\n",
    "            temp_data = pd.DataFrame(patient.loc[i] )\n",
    "            temp_data = temp_data.T\n",
    "            temp_data.columns = columns\n",
    "            temp_data = temp_data.reset_index()\n",
    "            temp_data = temp_data.drop('index', axis=1)\n",
    "            if i > 0:\n",
    "                data_row = data_row.join(temp_data)\n",
    "            else: \n",
    "                data_row = temp_data\n",
    "        \n",
    "        data_row['Mortality'] =label\n",
    "        data_row['Length'] = length\n",
    "        if counter == 0:\n",
    "            new_dataframe = data_row\n",
    "        else:\n",
    "            new_dataframe = new_dataframe.append(data_row)\n",
    "        counter=counter+1\n",
    "        if counter %500 == 0:\n",
    "            print(str(counter) + \"/\" + str(counter_length))\n",
    "    return new_dataframe\n",
    "            \n",
    "train_data = make_patients_to_rows(train_split)\n",
    "print(\"train done\")\n",
    "val_data = make_patients_to_rows(val_split)\n",
    "print(\"val done\")\n",
    "test_data = make_patients_to_rows(test_split)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.to_csv('./datasets/ihm/train_data_90.csv')\n",
    "test_data.to_csv('./datasets/ihm/test_data_90.csv')\n",
    "val_data.to_csv('./datasets/ihm/val_data_90.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}