eicu-data-paper.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# eICU Collaborative Research Database\n",
    "\n",
    "This notebook generates the tables and figures from the eICU Collaborative Research Database paper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/alistairewj/.local/lib/python3.5/site-packages/psycopg2/__init__.py:144: UserWarning: The psycopg2 wheel package will be renamed from release 2.8; in order to keep installing from binary please use \"pip install psycopg2-binary\" instead. For details see: <http://initd.org/psycopg/docs/install.html#binary-install-from-pypi>.\n",
      "  \"\"\")\n"
     ]
    }
   ],
   "source": [
    "from __future__ import print_function\n",
    "\n",
    "# Import libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import psycopg2\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import tableone\n",
    "\n",
    "# print out pretty pandas dataframes\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "from collections import OrderedDict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot settings\n",
    "%matplotlib inline\n",
    "plt.style.use('ggplot')\n",
    "fontsize = 20 # size for x and y ticks\n",
    "plt.rcParams['legend.fontsize'] = fontsize\n",
    "plt.rcParams.update({'font.size': fontsize})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Database config\n",
    "sqluser = 'postgres'\n",
    "dbname = 'eicu'\n",
    "schema_name = 'eicu_crd'\n",
    "sqlhost = 'localhost'\n",
    "sqlport = 5647"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Connect to the database\n",
    "con = psycopg2.connect(dbname=dbname, user=sqluser, host=sqlhost, port=sqlport)\n",
    "query_schema = 'SET search_path to public,' + schema_name + ';'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Display list of tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>tablename</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>admissiondrug</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>admissiondx</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>allergy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>apacheapsvar</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>apachepatientresult</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>apachepredvar</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>careplancareprovider</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>careplaneol</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>careplangeneral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>careplangoal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>careplaninfectiousdisease</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>customlab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>diagnosis</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>hospital</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>infusiondrug</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>intakeoutput</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>lab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>medication</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>microlab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>note</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>nurseassessment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>nursecare</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>nursecharting</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>pasthistory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>patient</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>physicalexam</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>respiratorycare</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>respiratorycharting</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>treatment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>vitalaperiodic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>vitalperiodic</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    tablename\n",
       "0               admissiondrug\n",
       "1                 admissiondx\n",
       "2                     allergy\n",
       "3                apacheapsvar\n",
       "4         apachepatientresult\n",
       "5               apachepredvar\n",
       "6        careplancareprovider\n",
       "7                 careplaneol\n",
       "8             careplangeneral\n",
       "9                careplangoal\n",
       "10  careplaninfectiousdisease\n",
       "11                  customlab\n",
       "12                  diagnosis\n",
       "13                   hospital\n",
       "14               infusiondrug\n",
       "15               intakeoutput\n",
       "16                        lab\n",
       "17                 medication\n",
       "18                   microlab\n",
       "19                       note\n",
       "20            nurseassessment\n",
       "21                  nursecare\n",
       "22              nursecharting\n",
       "23                pasthistory\n",
       "24                    patient\n",
       "25               physicalexam\n",
       "26            respiratorycare\n",
       "27        respiratorycharting\n",
       "28                  treatment\n",
       "29             vitalaperiodic\n",
       "30              vitalperiodic"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query = \\\n",
    "\"\"\"\n",
    "SELECT tablename \n",
    "FROM pg_catalog.pg_tables \n",
    "WHERE schemaname like 'eicu_crd'\n",
    "ORDER BY tablename;\n",
    "\"\"\"\n",
    "\n",
    "list_of_tables = pd.read_sql_query(query,con)\n",
    "list_of_tables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Demographics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "select\n",
    "    pt.PATIENTUNITSTAYID\n",
    "  , pt.patienthealthsystemstayid\n",
    "  , pt.uniquepid\n",
    "  , pt.hospitalid\n",
    "  , pt.wardid\n",
    "\n",
    "  -- **************************** --\n",
    "  -- ******* DEMOGRAPHICS ******* --\n",
    "  -- **************************** --\n",
    "  \n",
    "  -- patient features\n",
    "  , pt.gender\n",
    "  , pt.age\n",
    "  , case\n",
    "      when pt.age = '> 89' then 90\n",
    "      when pt.age = '' then NULL\n",
    "    else cast(pt.age as NUMERIC)\n",
    "    end as age_numeric\n",
    "  , pt.ethnicity\n",
    "\n",
    "  -- hospital level features\n",
    "  , hp.numbedscategory as hospital_size\n",
    "  , hp.teachingstatus as hospital_teaching_status\n",
    "  , hp.region as hospital_region\n",
    "  \n",
    "  -- patient hospital stay features\n",
    "  , (hospitaldischargeoffset - hospitaladmitoffset) / 60.0 / 24.0 as hospital_los\n",
    "  , pt.hospitaldischargelocation\n",
    "  , pt.hospitaldischargestatus -- 'Expired', 'Alive', ''\n",
    "  , apv.diedinhospital\n",
    "  \n",
    "  , pt.hospitaldischargeyear\n",
    "\n",
    "  -- ICU length of stay in days\n",
    "  , pt.unitdischargeoffset / 60.0 / 24.0 as unit_los\n",
    "  , pt.unittype\n",
    "  , pt.unitadmitsource\n",
    "  , pt.unitdischargelocation\n",
    "  , pt.unitdischargestatus\n",
    "\n",
    "  -- APACHE covariates for discharge location, etc\n",
    "  , apv.admitsource\n",
    "  , apv.dischargelocation\n",
    "  , apv.bedcount\n",
    "  , apv.readmit\n",
    "\n",
    "  -- APACHE IVa\n",
    "  , ag.apachedxgroup\n",
    "  , pt.apacheadmissiondx\n",
    "\n",
    "  -- apache comorbidities\n",
    "  , aav.dialysis\n",
    "  , apv.aids\n",
    "  , apv.hepaticfailure\n",
    "  , apv.cirrhosis\n",
    "  , apv.diabetes\n",
    "  , apv.immunosuppression\n",
    "  , apv.leukemia\n",
    "  , apv.lymphoma\n",
    "  , apv.metastaticcancer\n",
    "  , apv.thrombolytics\n",
    "\n",
    "  , pt.admissionheight -- height in cm\n",
    "  , pt.admissionweight -- weight in kg\n",
    "\n",
    "  -- ********************** --\n",
    "  -- *** APACHE APS VAR *** --\n",
    "  -- ********************** --\n",
    "\n",
    "  , aav.eyes\n",
    "  , aav.motor\n",
    "  , aav.verbal\n",
    "  , aav.eyes + aav.motor + aav.verbal as GCS\n",
    "  , aav.meds as UnableGCS\n",
    "  , aav.urine\n",
    "  , aav.pao2\n",
    "  , aav.fio2\n",
    "  , case when aav.pao2 = -1 or aav.fio2 = -1\n",
    "      then -1\n",
    "    else\n",
    "      aav.pao2 / aav.fio2 * 100\n",
    "    end as pao2fio2_apache\n",
    "\n",
    "\n",
    "  -- APACHE vital signs\n",
    "  , aav.temperature\n",
    "  , aav.respiratoryrate\n",
    "  , aav.heartrate\n",
    "  , aav.meanbp\n",
    "\n",
    "  -- APACHE labs\n",
    "  -- Noteably, APACHE does not include lactate, platelets, INR, WBC, or PTT\n",
    "  , aav.albumin\n",
    "  , aav.bilirubin\n",
    "  , aav.bun\n",
    "  , aav.creatinine\n",
    "  , aav.glucose\n",
    "  , aav.hematocrit\n",
    "  , aav.sodium\n",
    "  , aav.pco2\n",
    "  , aav.ph\n",
    "  , aav.intubated\n",
    "  , aav.wbc\n",
    "\n",
    "  -- *********************** --\n",
    "  -- *** APACHE PRED VAR *** --\n",
    "  -- *********************** --\n",
    "\n",
    "  -- Mechanical ventilation start and stop times are interfaced into\n",
    "  -- or documented directly into the respiratory flow sheet\n",
    "  -- or by updating the care plan in eCareManager.\n",
    "  -- however, we decided not to use flow sheet, but to use APACHE vent variable.\n",
    "\n",
    "  , apv.oobIntubDay1 as oobIntubDay1\n",
    "  , apv.oobVentDay1 as oobVentDay1\n",
    "  , apv.VENTDAY1 as VentDay1\n",
    "\n",
    "  -- APACHE PATIENT RESULT --\n",
    "  , apr.physicianSpeciality as physicianSpeciality\n",
    "  , apr.acutePhysiologyScore as acutePhysiologyScore\n",
    "  , apr.apacheScore as apacheScore\n",
    "  , apr.predictedICUMortality as predictedICUMortality\n",
    "  , apr.predictedICULOS as predictedICULOS\n",
    "  , apr.predictedHospitalMortality as predictedHospitalMortality\n",
    "  , apr.predictedHospitalLOS as predictedHospitalLOS\n",
    "  , apr.preopMI as preopMI\n",
    "  , apr.preopCardiacCath as preopCardiacCath\n",
    "  , apr.PTCAwithin24h as PTCAwithin24h\n",
    "  , apv.graftCount\n",
    "\n",
    "from patient pt\n",
    "left join hospital hp\n",
    "  on pt.hospitalid = hp.hospitalid\n",
    "left join apachepredvar apv\n",
    "  on pt.patientunitstayid = apv.patientunitstayid\n",
    "left join apacheapsvar aav\n",
    "  on pt.patientunitstayid = aav.patientunitstayid\n",
    "left join apachepatientresult apr\n",
    "  on pt.patientunitstayid = apr.patientunitstayid\n",
    "  and apr.apacheversion = 'IVa'\n",
    "left join apache_groups ag\n",
    "  on pt.patientunitstayid = ag.patientunitstayid\n",
    "\"\"\"\n",
    "\n",
    "df = pd.read_sql_query(query,con)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "200859 unit stays.\n",
      "166355 hospital stays.\n",
      "139367 patients.\n",
      "   208 hospitals.\n",
      "   335 units (wards).\n"
     ]
    }
   ],
   "source": [
    "# patient counts\n",
    "N_STAYS = df['patientunitstayid'].nunique()\n",
    "# note that the number of unique patientunitstayid == df.shape[0]\n",
    "print('{:6d} unit stays.'.format(N_STAYS))\n",
    "print('{:6d} hospital stays.'.format(df['patienthealthsystemstayid'].nunique()))\n",
    "print('{:6d} patients.'.format(df['uniquepid'].nunique()))\n",
    "print('{:6d} hospitals.'.format(df['hospitalid'].nunique()))\n",
    "print('{:6d} units (wards).'.format(df['wardid'].nunique()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# collapse some categories\n",
    "df['gender_grouped'] = df['gender']\n",
    "df.loc[ df['gender_grouped'] == 'Other', 'gender_grouped'] = 'Other or Unknown'\n",
    "df.loc[ df['gender_grouped'] == 'Unknown', 'gender_grouped'] = 'Other or Unknown'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical = [u'gender_grouped',\n",
    "               u'ethnicity', u'hospital_size', u'hospital_teaching_status',\n",
    "               u'hospital_region', \n",
    "               u'hospitaldischargeyear',\n",
    "               u'unittype',\n",
    "               # don't include admit/disch locations\n",
    "               # u'unitadmitsource', u'unitdischargelocation',\n",
    "               # u'hospitaldischargelocation',\n",
    "               \n",
    "               #  below are from apache table\n",
    "               # u'admitsource', u'dischargelocation',\n",
    "               u'unitdischargestatus', \n",
    "               u'hospitaldischargestatus'\n",
    "               #u'readmit', \n",
    "               # too many categories\n",
    "               # u'apacheadmissiondx', u'physicianspeciality',\n",
    "               # below are infrequently documented\n",
    "               #u'preopmi', u'preopcardiaccath', u'ptcawithin24h', u'graftcount'\n",
    "              ]\n",
    "\n",
    "# columns we want to include\n",
    "columns = ['age_numeric'\n",
    "           , u'unit_los', u'hospital_los'\n",
    "           , u'diedinhospital'\n",
    "           , u'admissionheight', u'admissionweight'\n",
    "          ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/alistairewj/.virtualenvs/eicu/lib/python3.5/site-packages/modality.py:930: RuntimeWarning: divide by zero encountered in double_scalars\n",
      "  k_start = (yU[1]-yU[0])/(xU[1]-xU[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{llll}\n",
      "\\toprule\n",
      "                &       & isnull &              overall \\\\\n",
      "variable & level &        &                      \\\\\n",
      "\\midrule\n",
      "n & {} &        &               200859 \\\\\n",
      "age\\_numeric & {} &     95 &  65.00 [53.00,76.00] \\\\\n",
      "unit\\_los & {} &      0 &     1.57 [0.82,2.97] \\\\\n",
      "hospital\\_los & {} &      0 &    5.49 [2.90,10.04] \\\\\n",
      "diedinhospital & {} &  29682 &          0.09 (0.29) \\\\\n",
      "admissionheight & {} &   4215 &       169.25 (13.69) \\\\\n",
      "admissionweight & {} &  16718 &        83.93 (27.09) \\\\\n",
      "gender\\_grouped & {} &      0 &           134 (0.07) \\\\\n",
      "                & Female &        &        92303 (45.95) \\\\\n",
      "                & Male &        &       108379 (53.96) \\\\\n",
      "                & Other or Unknown &        &            43 (0.02) \\\\\n",
      "ethnicity & {} &      0 &          2290 (1.14) \\\\\n",
      "                & African American &        &        21308 (10.61) \\\\\n",
      "                & Asian &        &          3270 (1.63) \\\\\n",
      "                & Caucasian &        &       155285 (77.31) \\\\\n",
      "                & Hispanic &        &          7464 (3.72) \\\\\n",
      "                & Native American &        &          1700 (0.85) \\\\\n",
      "                & Other/Unknown &        &          9542 (4.75) \\\\\n",
      "hospital\\_size & 100 - 249 &  25279 &         41966 (23.9) \\\\\n",
      "                & 250 - 499 &        &        45716 (26.04) \\\\\n",
      "                & <100 &        &         12593 (7.17) \\\\\n",
      "                & >= 500 &        &        75305 (42.89) \\\\\n",
      "hospital\\_teaching\\_status & False &      0 &       149181 (74.27) \\\\\n",
      "                & True &        &        51678 (25.73) \\\\\n",
      "hospital\\_region & Midwest &  13838 &        65950 (35.26) \\\\\n",
      "                & Northeast &        &         14429 (7.72) \\\\\n",
      "                & South &        &        60294 (32.24) \\\\\n",
      "                & West &        &        46348 (24.78) \\\\\n",
      "hospitaldischargeyear & 2014 &      0 &        95513 (47.55) \\\\\n",
      "                & 2015 &        &       105346 (52.45) \\\\\n",
      "unittype & CCU-CTICU &      0 &         15290 (7.61) \\\\\n",
      "                & CSICU &        &          9625 (4.79) \\\\\n",
      "                & CTICU &        &          6158 (3.07) \\\\\n",
      "                & Cardiac ICU &        &         12467 (6.21) \\\\\n",
      "                & MICU &        &          17465 (8.7) \\\\\n",
      "                & Med-Surg ICU &        &       113222 (56.37) \\\\\n",
      "                & Neuro ICU &        &         14451 (7.19) \\\\\n",
      "                & SICU &        &         12181 (6.06) \\\\\n",
      "unitdischargestatus & {} &      0 &            34 (0.02) \\\\\n",
      "                & Alive &        &       189918 (94.55) \\\\\n",
      "                & Expired &        &         10907 (5.43) \\\\\n",
      "hospitaldischargestatus & {} &      0 &          1751 (0.87) \\\\\n",
      "                & Alive &        &       181104 (90.16) \\\\\n",
      "                & Expired &        &         18004 (8.96) \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# output these demographics to a markdown file\n",
    "t1 = tableone.TableOne(df, columns=columns + categorical, categorical=categorical,\n",
    "                       nonnormal=['age_numeric', 'unit_los', 'hospital_los'])\n",
    "print(t1.to_latex())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of diagnoses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{ll}\n",
      "\\toprule\n",
      "{} &      overall \\\\\n",
      "level                                              &              \\\\\n",
      "\\midrule\n",
      "Sepsis, pulmonary                                  &  6823 (5.01) \\\\\n",
      "Infarction, acute myocardial (MI)                  &  5919 (4.34) \\\\\n",
      "CVA, cerebrovascular accident/stroke               &  5284 (3.88) \\\\\n",
      "CHF, congestive heart failure                      &  4840 (3.55) \\\\\n",
      "Sepsis, renal/UTI (including bladder)              &  4284 (3.14) \\\\\n",
      "Diabetic ketoacidosis                              &  4001 (2.94) \\\\\n",
      "CABG alone, coronary artery bypass grafting        &  3635 (2.67) \\\\\n",
      "Rhythm disturbance (atrial, supraventricular)      &  3474 (2.55) \\\\\n",
      "Cardiac arrest (with or without respiratory arr... &  3377 (2.48) \\\\\n",
      "Emphysema/bronchitis                               &  3304 (2.43) \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# diagnoses\n",
    "idxKeep = (~df['predictedhospitalmortality'].isnull()) & (df['predictedhospitalmortality'] != '-1')\n",
    "\n",
    "t1dx = tableone.TableOne(df.loc[idxKeep,['apacheadmissiondx']],\n",
    "                         categorical=['apacheadmissiondx'],\n",
    "                        limit=10)\n",
    "print(t1dx.tableone.loc['apacheadmissiondx', 'overall'].to_latex())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of APACHE diagnoses grouped by clinical meaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sepsis                                   & 18087 (16.40) \\\\\n",
      "Cerebrovascular accident                 &  9758 (8.85) \\\\\n",
      "Cardiac Arrest                           &  9135 (8.28) \\\\\n",
      "Acute Coronary Syndrome                  &  8343 (7.57) \\\\\n",
      "Respiratory medicine                     &  7970 (7.23) \\\\\n",
      "Gastrointestinal Bleed                   &  7277 (6.60) \\\\\n",
      "Congestive Heart Failure                 &  5884 (5.34) \\\\\n",
      "Trauma                                   &  5592 (5.07) \\\\\n",
      "Coronary Artery Bypass Graft             &  4771 (4.33) \\\\\n",
      "Neurological                             &  4640 (4.21) \\\\\n",
      "Pneumonia                                &  4577 (4.15) \\\\\n",
      "Diabetic Ketoacidosis                    &  4384 (3.98) \\\\\n",
      "Overdose                                 &  4268 (3.87) \\\\\n",
      "Asthma/Emphysema                         &  3948 (3.58) \\\\\n",
      "Other cardiovascular disease             &  3593 (3.26) \\\\\n",
      "Valvular disorders                       &  2795 (2.53) \\\\\n",
      "Coma                                     &  2082 (1.89) \\\\\n",
      "Acute renal failure                      &  1932 (1.75) \\\\\n",
      "Gastrointestinal obstruction             &  1232 (1.12) \\\\\n"
     ]
    }
   ],
   "source": [
    "pretty_names = {'Sepsis': 'Sepsis'\n",
    ", 'CVA': 'Cerebrovascular accident'\n",
    ", 'CardiacArrest': 'Cardiac Arrest'\n",
    ", 'ACS': 'Acute Coronary Syndrome'\n",
    ", 'RespMedOther': 'Respiratory medicine'\n",
    ", 'GIBleed': 'Gastrointestinal Bleed'\n",
    ", 'Trauma': 'Congestive Heart Failure'\n",
    ", 'CHF': 'Trauma'\n",
    ", 'CABG': 'Coronary Artery Bypass Graft'\n",
    ", 'Neuro': 'Neurological'\n",
    ", 'PNA': 'Pneumonia'\n",
    ", 'DKA': 'Diabetic Ketoacidosis'\n",
    ", 'Overdose': 'Overdose'\n",
    ", 'Asthma-Emphys': 'Asthma/Emphysema'\n",
    ", 'CVOther': 'Other cardiovascular disease'\n",
    ", 'ValveDz': 'Valvular disorders'\n",
    ", 'Coma': 'Coma'\n",
    ", 'ARF': 'Acute renal failure'\n",
    ", 'GIObstruction': 'Gastrointestinal obstruction'\n",
    ", 'ChestPain': 'Chest Pain (unknown cause)'}\n",
    "\n",
    "# diagnoses for only patients with APACHE predictions\n",
    "idxKeep = (~df['predictedhospitalmortality'].isnull()) & (df['predictedhospitalmortality'] != '-1')\n",
    "\n",
    "df_adx = df.loc[idxKeep, ['patientunitstayid','apachedxgroup']].copy()\n",
    "df_adx['apachedxgroup'] = df_adx['apachedxgroup'].map(pretty_names)\n",
    "df_adx = df_adx.groupby('apachedxgroup').count()\n",
    "df_adx['%'] = np.round(df_adx['patientunitstayid']*100.0/(df_adx['patientunitstayid'].sum()),2)\n",
    "\n",
    "for i, row in df_adx.sort_values('%', ascending=False).iterrows():\n",
    "    print('{:40s} & {:5g} ({:3.2f}) \\\\\\\\'.format(row.name, row['patientunitstayid'], np.round(row['%'],2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "200859 - 136236 = 64623 missing a diagnosis\n"
     ]
    },
    {
     "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></th>\n",
       "      <th>isnull</th>\n",
       "      <th>overall</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variable</th>\n",
       "      <th>level</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>n</th>\n",
       "      <th></th>\n",
       "      <td></td>\n",
       "      <td>136236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">apachedxgroup</th>\n",
       "      <th>ACS</th>\n",
       "      <td>0</td>\n",
       "      <td>8343 (6.12)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ARF</th>\n",
       "      <td></td>\n",
       "      <td>1932 (1.42)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Asthma-Emphys</th>\n",
       "      <td></td>\n",
       "      <td>3948 (2.9)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CABG</th>\n",
       "      <td></td>\n",
       "      <td>4771 (3.5)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            isnull      overall\n",
       "variable      level                            \n",
       "n                                        136236\n",
       "apachedxgroup ACS                0  8343 (6.12)\n",
       "              ARF                   1932 (1.42)\n",
       "              Asthma-Emphys          3948 (2.9)\n",
       "              CABG                   4771 (3.5)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# diagnoses for only patients with APACHE predictions\n",
    "idxKeep = (~df['predictedhospitalmortality'].isnull()) & (df['predictedhospitalmortality'] != '-1')\n",
    "t1dx = tableone.TableOne(df.loc[idxKeep,:], columns=['apachedxgroup'], categorical=['apachedxgroup'])\n",
    "print('{} - {} = {} missing a diagnosis'.format(df.shape[0], np.sum(idxKeep), df.shape[0]-np.sum(idxKeep)))\n",
    "t1dx.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of hospital stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "-- hospital level features\n",
    "select\n",
    "    hp.hospitalid\n",
    "  , hp.numbedscategory as hospital_size\n",
    "  , hp.teachingstatus as hospital_teaching_status\n",
    "  , hp.region as hospital_region\n",
    "  , count(distinct pt.patientunitstayid) as n_patients\n",
    "from hospital hp\n",
    "inner join patient pt\n",
    "  on hp.hospitalid = pt.hospitalid\n",
    "group by hp.hospitalid, hp.numbedscategory, hp.teachingstatus, hp.region\n",
    "\"\"\"\n",
    "\n",
    "df_hosp = pd.read_sql_query(query,con)\n",
    "\n",
    "# set null to unknown category so tableone uses desired format\n",
    "df_hosp['hospital_size'].fillna('Unknown',inplace=True)\n",
    "df_hosp['hospital_teaching_status'].fillna('Unknown',inplace=True)\n",
    "df_hosp['hospital_region'].fillna('Unknown',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hospitalid</th>\n",
       "      <th>hospitalid_%</th>\n",
       "      <th>n_patients</th>\n",
       "      <th>n_patients_%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hospital_size</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>100 - 249</th>\n",
       "      <td>62</td>\n",
       "      <td>29.81</td>\n",
       "      <td>41966</td>\n",
       "      <td>20.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250 - 499</th>\n",
       "      <td>35</td>\n",
       "      <td>16.83</td>\n",
       "      <td>45716</td>\n",
       "      <td>22.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>&lt;100</th>\n",
       "      <td>46</td>\n",
       "      <td>22.12</td>\n",
       "      <td>12593</td>\n",
       "      <td>6.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>&gt;= 500</th>\n",
       "      <td>23</td>\n",
       "      <td>11.06</td>\n",
       "      <td>75305</td>\n",
       "      <td>37.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unknown</th>\n",
       "      <td>42</td>\n",
       "      <td>20.19</td>\n",
       "      <td>25279</td>\n",
       "      <td>12.59</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hospitalid</th>\n",
       "      <th>hospitalid_%</th>\n",
       "      <th>n_patients</th>\n",
       "      <th>n_patients_%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hospital_teaching_status</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>189</td>\n",
       "      <td>90.87</td>\n",
       "      <td>149181</td>\n",
       "      <td>74.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>19</td>\n",
       "      <td>9.13</td>\n",
       "      <td>51678</td>\n",
       "      <td>25.73</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hospitalid</th>\n",
       "      <th>hospitalid_%</th>\n",
       "      <th>n_patients</th>\n",
       "      <th>n_patients_%</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hospital_region</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Midwest</th>\n",
       "      <td>70</td>\n",
       "      <td>33.65</td>\n",
       "      <td>65950</td>\n",
       "      <td>32.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Northeast</th>\n",
       "      <td>13</td>\n",
       "      <td>6.25</td>\n",
       "      <td>14429</td>\n",
       "      <td>7.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>South</th>\n",
       "      <td>56</td>\n",
       "      <td>26.92</td>\n",
       "      <td>60294</td>\n",
       "      <td>30.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unknown</th>\n",
       "      <td>26</td>\n",
       "      <td>12.50</td>\n",
       "      <td>13838</td>\n",
       "      <td>6.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>West</th>\n",
       "      <td>43</td>\n",
       "      <td>20.67</td>\n",
       "      <td>46348</td>\n",
       "      <td>23.07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for h in ['hospital_size','hospital_teaching_status','hospital_region']:\n",
    "    grp = df_hosp.groupby(h)[['hospitalid','n_patients']].agg({'hospitalid': 'count', 'n_patients': 'sum'})\n",
    "    grp['hospitalid_%'] = np.round(grp['hospitalid'] / grp['hospitalid'].sum() * 100.0,2)\n",
    "    grp['n_patients_%'] = np.round(grp['n_patients'] / grp['n_patients'].sum() * 100.0,2)\n",
    "    display(HTML(grp[['hospitalid', 'hospitalid_%', 'n_patients','n_patients_%']].to_html()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Information on top diagnoses/treatments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3933, 3) unique diagnoses.\n"
     ]
    },
    {
     "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>diagnosisstring</th>\n",
       "      <th>numobs</th>\n",
       "      <th>frac_pat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>pulmonary|respiratory failure|acute respirator...</td>\n",
       "      <td>22395</td>\n",
       "      <td>11.149612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>renal|disorder of kidney|acute renal failure</td>\n",
       "      <td>16375</td>\n",
       "      <td>8.152485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>endocrine|glucose metabolism|diabetes mellitus</td>\n",
       "      <td>14620</td>\n",
       "      <td>7.278738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cardiovascular|ventricular disorders|congestiv...</td>\n",
       "      <td>12819</td>\n",
       "      <td>6.382089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>pulmonary|pulmonary infections|pneumonia</td>\n",
       "      <td>12347</td>\n",
       "      <td>6.147098</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     diagnosisstring  numobs   frac_pat\n",
       "0  pulmonary|respiratory failure|acute respirator...   22395  11.149612\n",
       "1       renal|disorder of kidney|acute renal failure   16375   8.152485\n",
       "2     endocrine|glucose metabolism|diabetes mellitus   14620   7.278738\n",
       "3  cardiovascular|ventricular disorders|congestiv...   12819   6.382089\n",
       "4           pulmonary|pulmonary infections|pneumonia   12347   6.147098"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# diagnoses\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "select diagnosisstring\n",
    ", count(distinct patientunitstayid) as numobs\n",
    "from diagnosis\n",
    "group by diagnosisstring\n",
    "order by numobs desc\n",
    "\"\"\"\n",
    "df_misc = pd.read_sql_query(query,con)\n",
    "\n",
    "df_misc['frac_pat'] = df_misc['numobs']/200859*100.0\n",
    "print('{} unique diagnoses.'.format(df_misc.shape))\n",
    "display(df_misc.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2711, 3) unique treatments.\n",
      " 34059 (16.96) - pulmonary|ventilation and oxygenation|mechanical ventilation\n",
      " 17660 ( 8.79) - pulmonary|radiologic procedures / bronchoscopy|chest x-ray\n",
      " 15200 ( 7.57) - cardiovascular|intravenous fluid|normal saline administration\n",
      " 13926 ( 6.93) - pulmonary|ventilation and oxygenation|oxygen therapy (< 40%)|nasal cannula\n",
      " 12424 ( 6.19) - infectious diseases|medications|therapeutic antibacterials\n",
      " 12383 ( 6.17) - pulmonary|ventilation and oxygenation|non-invasive ventilation\n"
     ]
    }
   ],
   "source": [
    "# treatments\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "select treatmentstring\n",
    ", count(distinct patientunitstayid) as numobs\n",
    "from treatment\n",
    "group by treatmentstring\n",
    "order by numobs desc\n",
    "\"\"\"\n",
    "df_treat = pd.read_sql_query(query,con)\n",
    "\n",
    "df_treat['frac_pat'] = df_treat['numobs']/200859*100.0\n",
    "print('{} unique treatments.'.format(df_treat.shape))\n",
    "\n",
    "for i, row in df_treat.iterrows():\n",
    "    print('{:6g} ({:5.2f}) - {}'.format(\n",
    "            row['numobs'], row['frac_pat'], row['treatmentstring']))\n",
    "    if i>=5:\n",
    "        break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of top diagnosis groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cardiovascular                           & 104,264 (11.15\\%) \\\\\n",
      "Pulmonary                                & 64,222 (8.15\\%) \\\\\n",
      "Neurologic                               & 51,609 (7.28\\%) \\\\\n",
      "Renal                                    & 43,009 (6.38\\%) \\\\\n",
      "Endocrine                                & 35,519 (6.15\\%) \\\\\n",
      "Gastrointestinal                         & 35,223 (6.10\\%) \\\\\n",
      "Infectious diseases                      & 20,316 (6.01\\%) \\\\\n",
      "Hematology                               & 19,611 (5.32\\%) \\\\\n",
      "Burns/trauma                             & 9,208 (5.13\\%) \\\\\n",
      "Oncology                                 & 7,954 (4.72\\%) \\\\\n",
      "Toxicology                               & 7,185 (4.47\\%) \\\\\n",
      "Surgery                                  & 5,723 (3.97\\%) \\\\\n",
      "General                                  & 1,698 (3.91\\%) \\\\\n",
      "Transplant                               &   770 (3.75\\%) \\\\\n",
      "Obstetrics/gynecology                    &    46 (3.52\\%) \\\\\n",
      "Genitourinary                            &    26 (3.18\\%) \\\\\n",
      "Musculoskeletal                          &    19 (2.98\\%) \\\\\n"
     ]
    }
   ],
   "source": [
    "# diagnosis groups\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "with t1 as\n",
    "(\n",
    "select substring(diagnosisstring,1,position('|' in diagnosisstring)-1) as dx\n",
    ", patientunitstayid\n",
    "from diagnosis\n",
    ")\n",
    "select dx, count(distinct patientunitstayid) as numpat\n",
    "from t1\n",
    "group by dx\n",
    "order by numpat desc\n",
    "\"\"\"\n",
    "df_dxgrp = pd.read_sql_query(query,con)\n",
    "\n",
    "df_dxgrp['frac_pat'] = df_misc['numobs']/200859*100.0\n",
    "for i, row in df_dxgrp.sort_values('numpat', ascending=False).iterrows():\n",
    "    print('{:40s} & {:5,d} ({:3.2f}\\%) \\\\\\\\'.format(row['dx'].capitalize(),\n",
    "                                                 row['numpat'],\n",
    "                                                 row['frac_pat']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Number of observations for vital tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "with t1 as\n",
    "(\n",
    "select patientunitstayid\n",
    ", count(heartrate) as heartrate\n",
    ", count(respiration) as respiration\n",
    ", count(sao2) as sao2\n",
    ", count(temperature) as temperature\n",
    ", count(cvp) as cvp\n",
    ", count(etco2) as etco2\n",
    ", count(systemicsystolic) as systemicsystolic\n",
    ", count(systemicdiastolic) as systemicdiastolic\n",
    ", count(systemicmean) as systemicmean\n",
    ", count(pasystolic) as pasystolic\n",
    ", count(padiastolic) as padiastolic\n",
    ", count(pamean) as pamean\n",
    ", count(st1) as st1\n",
    ", count(st2) as st2\n",
    ", count(st3) as st3\n",
    ", count(icp) as icp\n",
    "from vitalperiodic\n",
    "group by patientunitstayid\n",
    ")\n",
    "select\n",
    "  'patient count' as data_type\n",
    ", sum(case when heartrate > 0 then 1 else 0 end) as heartrate\n",
    ", sum(case when respiration > 0 then 1 else 0 end) as respiration\n",
    ", sum(case when sao2 > 0 then 1 else 0 end) as sao2\n",
    ", sum(case when temperature > 0 then 1 else 0 end) as temperature\n",
    ", sum(case when cvp > 0 then 1 else 0 end) as cvp\n",
    ", sum(case when etco2 > 0 then 1 else 0 end) as etco2\n",
    ", sum(case when systemicsystolic > 0 then 1 else 0 end) as systemicsystolic\n",
    ", sum(case when systemicdiastolic > 0 then 1 else 0 end) as systemicdiastolic\n",
    ", sum(case when systemicmean > 0 then 1 else 0 end) as systemicmean\n",
    ", sum(case when pasystolic > 0 then 1 else 0 end) as pasystolic\n",
    ", sum(case when padiastolic > 0 then 1 else 0 end) as padiastolic\n",
    ", sum(case when pamean > 0 then 1 else 0 end) as pamean\n",
    ", sum(case when st1 > 0 then 1 else 0 end) as st1\n",
    ", sum(case when st2 > 0 then 1 else 0 end) as st2\n",
    ", sum(case when st3 > 0 then 1 else 0 end) as st3\n",
    ", sum(case when icp > 0 then 1 else 0 end) as icp\n",
    "from t1\n",
    "UNION\n",
    "select\n",
    "  'observation count' as data_type\n",
    ", sum(heartrate)::numeric / sum(case when heartrate > 0 then 1 else 0 end) as heartrate\n",
    ", sum(respiration)::numeric / sum(case when respiration > 0 then 1 else 0 end) as respiration\n",
    ", sum(sao2)::numeric / sum(case when sao2 > 0 then 1 else 0 end) as sao2\n",
    ", sum(temperature)::numeric / sum(case when temperature > 0 then 1 else 0 end) as temperature\n",
    ", sum(cvp)::numeric / sum(case when cvp > 0 then 1 else 0 end) as cvp\n",
    ", sum(etco2)::numeric / sum(case when etco2 > 0 then 1 else 0 end) as etco2\n",
    ", sum(systemicsystolic)::numeric / sum(case when systemicsystolic > 0 then 1 else 0 end) as systemicsystolic\n",
    ", sum(systemicdiastolic)::numeric / sum(case when systemicdiastolic > 0 then 1 else 0 end) as systemicdiastolic\n",
    ", sum(systemicmean)::numeric / sum(case when systemicmean > 0 then 1 else 0 end) as systemicmean\n",
    ", sum(pasystolic)::numeric / sum(case when pasystolic > 0 then 1 else 0 end) as pasystolic\n",
    ", sum(padiastolic)::numeric / sum(case when padiastolic > 0 then 1 else 0 end) as padiastolic\n",
    ", sum(pamean)::numeric / sum(case when pamean > 0 then 1 else 0 end) as pamean\n",
    ", sum(st1)::numeric / sum(case when st1 > 0 then 1 else 0 end) as st1\n",
    ", sum(st2)::numeric / sum(case when st2 > 0 then 1 else 0 end) as st2\n",
    ", sum(st3)::numeric / sum(case when st3 > 0 then 1 else 0 end) as st3\n",
    ", sum(icp)::numeric / sum(case when icp > 0 then 1 else 0 end) as icp\n",
    "from t1;\n",
    "\"\"\"\n",
    "df_vp = pd.read_sql_query(query,con)\n",
    "df_vp.set_index('data_type',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>data_type</th>\n",
       "      <th>observation count</th>\n",
       "      <th>patient count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>heartrate</th>\n",
       "      <td>759.216100</td>\n",
       "      <td>192277.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>respiration</th>\n",
       "      <td>721.709128</td>\n",
       "      <td>178051.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sao2</th>\n",
       "      <td>700.822933</td>\n",
       "      <td>189646.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>temperature</th>\n",
       "      <td>679.916010</td>\n",
       "      <td>19419.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cvp</th>\n",
       "      <td>667.564221</td>\n",
       "      <td>28698.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>etco2</th>\n",
       "      <td>529.994369</td>\n",
       "      <td>8346.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>systemicsystolic</th>\n",
       "      <td>596.459168</td>\n",
       "      <td>46667.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>systemicdiastolic</th>\n",
       "      <td>596.512012</td>\n",
       "      <td>46661.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>systemicmean</th>\n",
       "      <td>597.357531</td>\n",
       "      <td>46975.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pasystolic</th>\n",
       "      <td>381.975901</td>\n",
       "      <td>10789.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>padiastolic</th>\n",
       "      <td>381.823202</td>\n",
       "      <td>10792.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pamean</th>\n",
       "      <td>380.990728</td>\n",
       "      <td>10893.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>st1</th>\n",
       "      <td>591.835440</td>\n",
       "      <td>95643.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>st2</th>\n",
       "      <td>606.246314</td>\n",
       "      <td>98886.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>st3</th>\n",
       "      <td>595.148773</td>\n",
       "      <td>92752.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>icp</th>\n",
       "      <td>1610.298042</td>\n",
       "      <td>1634.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "data_type          observation count  patient count\n",
       "heartrate                 759.216100       192277.0\n",
       "respiration               721.709128       178051.0\n",
       "sao2                      700.822933       189646.0\n",
       "temperature               679.916010        19419.0\n",
       "cvp                       667.564221        28698.0\n",
       "etco2                     529.994369         8346.0\n",
       "systemicsystolic          596.459168        46667.0\n",
       "systemicdiastolic         596.512012        46661.0\n",
       "systemicmean              597.357531        46975.0\n",
       "pasystolic                381.975901        10789.0\n",
       "padiastolic               381.823202        10792.0\n",
       "pamean                    380.990728        10893.0\n",
       "st1                       591.835440        95643.0\n",
       "st2                       606.246314        98886.0\n",
       "st3                       595.148773        92752.0\n",
       "icp                      1610.298042         1634.0"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vp.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Heart rate                               & \\colname{heartrate}          & 192,277 (95.73\\%) & 145,979,794 (759.2) \\\\\n",
      "Peripheral oxygen saturation             & \\colname{sao2}               & 189,646 (94.42\\%) & 132,908,266 (700.8) \\\\\n",
      "Respiration rate                         & \\colname{respiration}        & 178,051 (88.64\\%) & 128,501,032 (721.7) \\\\\n",
      "ST level                                 & \\colname{st2}                & 98,886 (49.23\\%) & 59,949,273 (606.2) \\\\\n",
      "ST level                                 & \\colname{st1}                & 95,643 (47.62\\%) & 56,604,917 (591.8) \\\\\n",
      "ST level                                 & \\colname{st3}                & 92,752 (46.18\\%) & 55,201,239 (595.1) \\\\\n",
      "Invasive mean blood pressure             & \\colname{systemicmean}       & 46,975 (23.39\\%) & 28,060,870 (597.4) \\\\\n",
      "Invasive systolic blood pressure         & \\colname{systemicsystolic}   & 46,667 (23.23\\%) & 27,834,959 (596.5) \\\\\n",
      "Invasive diastolic blood pressure        & \\colname{systemicdiastolic}  & 46,661 (23.23\\%) & 27,833,847 (596.5) \\\\\n",
      "Central venous pressure                  & \\colname{cvp}                & 28,698 (14.29\\%) & 19,157,758 (667.6) \\\\\n",
      "Temperature                              & \\colname{temperature}        & 19,419 (9.67\\%) & 13,203,289 (679.9) \\\\\n",
      "Mean pulmonary artery pressure           & \\colname{pamean}             & 10,893 (5.42\\%) & 4,150,132 (381.0) \\\\\n",
      "Diastolic pulmonary artery pressure      & \\colname{padiastolic}        & 10,792 (5.37\\%) & 4,120,636 (381.8) \\\\\n",
      "Systolic pulmonary artery pressure       & \\colname{pasystolic}         & 10,789 (5.37\\%) & 4,121,138 (382.0) \\\\\n",
      "End tidal carbon dioxide concentration   & \\colname{etco2}              &  8,346 (4.16\\%) & 4,423,333 (530.0) \\\\\n",
      "Intracranial pressure                    & \\colname{icp}                &  1,634 (0.81\\%) & 2,631,227 (1610.3) \\\\\n"
     ]
    }
   ],
   "source": [
    "pretty_labels = {'heartrate': 'Heart rate'\n",
    ", 'respiration': 'Respiration rate'\n",
    ", 'sao2': 'Peripheral oxygen saturation'\n",
    ", 'temperature': 'Temperature'\n",
    ", 'cvp': 'Central venous pressure'\n",
    ", 'etco2': 'End tidal carbon dioxide concentration'\n",
    ", 'systemicsystolic': 'Invasive systolic blood pressure'\n",
    ", 'systemicdiastolic': 'Invasive diastolic blood pressure'\n",
    ", 'systemicmean': 'Invasive mean blood pressure'\n",
    ", 'pasystolic': 'Systolic pulmonary artery pressure'\n",
    ", 'padiastolic': 'Diastolic pulmonary artery pressure'\n",
    ", 'pamean': 'Mean pulmonary artery pressure'\n",
    ", 'st1': 'ST level'\n",
    ", 'st2': 'ST level'\n",
    ", 'st3': 'ST level'\n",
    ", 'icp': 'Intracranial pressure'}\n",
    "\n",
    "df_vp.sort_values('patient count',axis=1,ascending=False,inplace=True)\n",
    "\n",
    "for c in df_vp.columns:\n",
    "    if c != 'data_type':\n",
    "        print('{:40s}'.format(pretty_labels[c]), end=' & ')\n",
    "        print('\\\\colname{:20s}'.format('{' + c + '}'),end=' & ')\n",
    "        print('{:6,d} ({:3.2f}\\%)'.format(\n",
    "                df_vp.loc['patient count',c].astype(int),\n",
    "                df_vp.loc['patient count',c].astype(float)/200859*100.0),\n",
    "              end=' & ')\n",
    "        print('{:,d} ({:3.1f})'.format((df_vp.loc['observation count',c]*df_vp.loc['patient count',c]).astype(int),\n",
    "                               df_vp.loc['observation count',c]),\n",
    "              end=' \\\\\\\\\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# vital aperiodic\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "with t1 as\n",
    "(\n",
    "select patientunitstayid\n",
    ", count(noninvasivesystolic) as noninvasivesystolic\n",
    ", count(noninvasivediastolic) as noninvasivediastolic\n",
    ", count(noninvasivemean) as noninvasivemean\n",
    ", count(paop) as paop\n",
    ", count(cardiacoutput) as cardiacoutput\n",
    ", count(cardiacinput) as cardiacinput\n",
    ", count(svr) as svr\n",
    ", count(svri) as svri\n",
    ", count(pvr) as pvr\n",
    ", count(pvri) as pvri\n",
    "from vitalaperiodic\n",
    "group by patientunitstayid\n",
    ")\n",
    "select\n",
    "  sum(case when noninvasivesystolic > 0 then 1 else 0 end) as noninvasivesystolic\n",
    ", sum(case when noninvasivediastolic > 0 then 1 else 0 end) as noninvasivediastolic\n",
    ", sum(case when noninvasivemean > 0 then 1 else 0 end) as noninvasivemean\n",
    ", sum(case when paop > 0 then 1 else 0 end) as paop\n",
    ", sum(case when cardiacoutput > 0 then 1 else 0 end) as cardiacoutput\n",
    ", sum(case when cardiacinput > 0 then 1 else 0 end) as cardiacinput\n",
    ", sum(case when svr > 0 then 1 else 0 end) as svr\n",
    ", sum(case when svri > 0 then 1 else 0 end) as svri\n",
    ", sum(case when pvr > 0 then 1 else 0 end) as pvr\n",
    ", sum(case when pvri > 0 then 1 else 0 end) as pvri\n",
    "from t1;\n",
    "\"\"\"\n",
    "df_vap = pd.read_sql_query(query,con)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>noninvasivesystolic</th>\n",
       "      <td>189620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>noninvasivediastolic</th>\n",
       "      <td>189626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>noninvasivemean</th>\n",
       "      <td>189663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>paop</th>\n",
       "      <td>2072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cardiacoutput</th>\n",
       "      <td>5591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cardiacinput</th>\n",
       "      <td>4203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>svr</th>\n",
       "      <td>5022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>svri</th>\n",
       "      <td>4156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pvr</th>\n",
       "      <td>1898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pvri</th>\n",
       "      <td>1875</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           0\n",
       "noninvasivesystolic   189620\n",
       "noninvasivediastolic  189626\n",
       "noninvasivemean       189663\n",
       "paop                    2072\n",
       "cardiacoutput           5591\n",
       "cardiacinput            4203\n",
       "svr                     5022\n",
       "svri                    4156\n",
       "pvr                     1898\n",
       "pvri                    1875"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_vap.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "189620 (94.40%) - noninvasivesystolic\n",
      "189626 (94.41%) - noninvasivediastolic\n",
      "189663 (94.43%) - noninvasivemean\n",
      "  2072 ( 1.03%) - paop\n",
      "  5591 ( 2.78%) - cardiacoutput\n",
      "  4203 ( 2.09%) - cardiacinput\n",
      "  5022 ( 2.50%) - svr\n",
      "  4156 ( 2.07%) - svri\n",
      "  1898 ( 0.94%) - pvr\n",
      "  1875 ( 0.93%) - pvri\n"
     ]
    }
   ],
   "source": [
    "for c in df_vap.columns:\n",
    "    print('{:6g} ({:5.2f}%) - {}'.format(\n",
    "            df_vap.loc[0,c],\n",
    "            df_vap.loc[0,c].astype(float)/200859*100.0,\n",
    "            c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Number of distinct laboratory measurements per patient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# query to load data from the patient table\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", COUNT(DISTINCT labname) AS num_distinct_labs\n",
    "FROM lab\n",
    "GROUP BY patientunitstayid\n",
    "\"\"\"\n",
    "\n",
    "df_lab = pd.read_sql_query(query, con)\n",
    "\n",
    "plt.figure(figsize=[12,8])\n",
    "plt.hist(df_lab['num_distinct_labs'], bins=np.linspace(0,160,33))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Funnel plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT pt.patientunitstayid, pt.hospitalid\n",
    ", apr.predictedhospitalmortality as pred\n",
    ", CASE WHEN pt.hospitaldischargestatus = 'Expired' then 1\n",
    "       WHEN pt.hospitaldischargestatus = 'Alive' then 0\n",
    "    else NULL end as target\n",
    "FROM patient pt\n",
    "INNER JOIN apachepatientresult apr\n",
    "ON pt.patientunitstayid = apr.patientunitstayid\n",
    "AND apr.apacheversion = 'IVa'\n",
    "WHERE apr.predictedhospitalmortality != ''\n",
    "AND apr.predictedhospitalmortality IS NOT NULL\n",
    "AND apr.predictedhospitalmortality != '-1'\n",
    "AND pt.hospitaldischargestatus in ('Alive','Expired')\n",
    "\"\"\"\n",
    "\n",
    "df = pd.read_sql_query(query,con)\n",
    "df['pred'] = pd.to_numeric(df['pred'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# count patients per hospital\n",
    "idx = df.groupby('hospitalid')['patientunitstayid'].count()\n",
    "# isolate to only those with count >= 100\n",
    "idx = idx.index[idx>=100]\n",
    "# create index for df\n",
    "idx = np.in1d(df['hospitalid'],idx)\n",
    "\n",
    "grp = df.loc[idx,:].groupby('hospitalid')\n",
    "\n",
    "plt.figure(figsize=[12,8])\n",
    "plt.plot(grp['patientunitstayid'].count(),\n",
    "         grp['pred'].mean() / grp['target'].mean(),\n",
    "        'o', markersize=8)\n",
    "\n",
    "plt.plot([0,5000],\n",
    "         np.ones(2) * np.mean(grp['pred'].mean() / grp['target'].mean()),\n",
    "        'k--', linewidth=2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data completion for patientunitstayid\n",
    "\n",
    "Below query returns whether a `patientUnitStayId` is in each table in the database."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# query to load data from the patient table\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT\n",
    "  pt.patientunitstayid, pt.hospitalid\n",
    "  , CASE WHEN adx.patientunitstayid  IS NOT NULL then 1 else 0 end as admissionDx\n",
    "  , CASE WHEN adr.patientunitstayid  IS NOT NULL then 1 else 0 end as admissionDrug\n",
    "  , CASE WHEN al.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS allergy\n",
    "  , CASE WHEN aav.patientunitstayid  IS NOT NULL then 1 else 0 end as apacheApsVar\n",
    "  , CASE WHEN apv.patientunitstayid  IS NOT NULL then 1 else 0 end as apachePredVar\n",
    "  , CASE WHEN apr.patientunitstayid  IS NOT NULL then 1 else 0 end as apachePatientResult\n",
    "  , CASE WHEN cpp.patientunitstayid  IS NOT NULL then 1 else 0 end as carePlanCareProvider\n",
    "  , CASE WHEN cpe.patientunitstayid  IS NOT NULL then 1 else 0 end as carePlanEOL\n",
    "  , CASE WHEN cpg.patientunitstayid  IS NOT NULL then 1 else 0 end as carePlanGeneral\n",
    "  , CASE WHEN cpgl.patientunitstayid IS NOT NULL then 1 else 0 end as carePlanGoal\n",
    "  , CASE WHEN cpi.patientunitstayid  IS NOT NULL then 1 else 0 end as carePlanInfectiousDisease\n",
    "  , CASE WHEN cl.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS customLab\n",
    "  , CASE WHEN di.patientunitstayid   IS NOT NULL then 1 else 0 end as diagnosis\n",
    "  , CASE WHEN id.patientunitstayid   IS NOT NULL then 1 else 0 end as infusionDrug\n",
    "  , CASE WHEN io.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS intakeOutput\n",
    "  , CASE WHEN la.patientunitstayid   IS NOT NULL then 1 else 0 end as lab\n",
    "  , CASE WHEN me.patientunitstayid   IS NOT NULL then 1 else 0 end as medication\n",
    "  , CASE WHEN mi.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS microLab\n",
    "  , CASE WHEN no.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS note\n",
    "  , CASE WHEN na.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS nurseAssessment\n",
    "  , CASE WHEN nca.patientunitstayid  IS NOT NULL THEN 1 ELSE 0 END AS nurseCare\n",
    "  , CASE WHEN nc.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS nurseCharting\n",
    "  , CASE WHEN pa.patientunitstayid   IS NOT NULL then 1 else 0 end as pastHistory\n",
    "  , CASE WHEN pe.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS physicalExam\n",
    "  , CASE WHEN rca.patientunitstayid  IS NOT NULL THEN 1 ELSE 0 END AS respiratoryCare\n",
    "  , CASE WHEN rc.patientunitstayid   IS NOT NULL THEN 1 ELSE 0 END AS respiratoryCharting\n",
    "  , CASE WHEN tr.patientunitstayid   IS NOT NULL then 1 else 0 end as treatment\n",
    "  , CASE WHEN va.patientunitstayid   IS NOT NULL then 1 else 0 end as vitalAperiodic\n",
    "  , CASE WHEN vp.patientunitstayid   IS NOT NULL then 1 else 0 end as vitalPeriodic\n",
    "FROM patient pt\n",
    "LEFT JOIN (select distinct patientunitstayid from admissiondx) adx\n",
    "    ON pt.patientunitstayid = adx.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from admissionDrug) adr\n",
    "    ON pt.patientunitstayid = adr.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from allergy) al\n",
    "    ON pt.patientunitstayid = al.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from apacheapsvar) aav\n",
    "    ON pt.patientunitstayid = aav.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from apachepredvar) apv\n",
    "    ON pt.patientunitstayid = apv.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from apachepatientresult) apr\n",
    "    ON pt.patientunitstayid = apr.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from careplancareprovider) cpp\n",
    "    ON pt.patientunitstayid = cpp.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from careplaneol) cpe\n",
    "    ON pt.patientunitstayid = cpe.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from careplangeneral) cpg\n",
    "    ON pt.patientunitstayid = cpg.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from careplangoal) cpgl\n",
    "    ON pt.patientunitstayid = cpgl.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from careplaninfectiousdisease) cpi\n",
    "    ON pt.patientunitstayid = cpi.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from customlab) cl\n",
    "    ON pt.patientunitstayid = cl.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from diagnosis) di\n",
    "    ON pt.patientunitstayid = di.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from infusionDrug) id\n",
    "    ON pt.patientunitstayid = id.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from intakeoutput) io\n",
    "    ON pt.patientunitstayid = io.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from lab) la\n",
    "    ON pt.patientunitstayid = la.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from medication) me\n",
    "    ON pt.patientunitstayid = me.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from microlab) mi\n",
    "    ON pt.patientunitstayid = mi.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from note) no\n",
    "    ON pt.patientunitstayid = no.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from nurseassessment) na\n",
    "    ON pt.patientunitstayid = na.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from nursecare) nca\n",
    "    ON pt.patientunitstayid = nca.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from nursecharting) nc\n",
    "    ON pt.patientunitstayid = nc.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from pasthistory) pa\n",
    "    ON pt.patientunitstayid = pa.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from physicalexam) pe\n",
    "    ON pt.patientunitstayid = pe.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from respiratorycare) rca\n",
    "    ON pt.patientunitstayid = rca.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from respiratorycharting) rc\n",
    "    ON pt.patientunitstayid = rc.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from treatment) tr\n",
    "    ON pt.patientunitstayid = tr.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from vitalaperiodic) va\n",
    "    ON pt.patientunitstayid = va.patientunitstayid\n",
    "LEFT JOIN (select distinct patientunitstayid from vitalperiodic) vp\n",
    "    ON pt.patientunitstayid = vp.patientunitstayid;\n",
    "\"\"\"\n",
    "\n",
    "df_dc = pd.read_sql_query(query,con)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# group by hospital - percent data completion for each table\n",
    "grp = df_dc.groupby('hospitalid')\n",
    "\n",
    "# group into 5 categories:\n",
    "# 0%, 0-20%, 20-60%, 60-80%, 80-100%\n",
    "lower_range = [ -1,  0, 20, 60,  80]\n",
    "upper_range = [  0, 20, 60, 80, 100]\n",
    "grp = grp.mean()*100.0\n",
    "num_hosp = grp.shape[0]\n",
    "\n",
    "df_count = pd.DataFrame(columns=df_dc.columns)\n",
    "df_count.drop(['patientunitstayid', 'hospitalid'], axis=1, inplace=True)\n",
    "for i in range(len(upper_range)):\n",
    "    df_count.loc[i,:] = None\n",
    "    for c in df_count.columns:\n",
    "        num_hosp_in_range=np.sum(\n",
    "            (grp[c] > lower_range[i]) & (grp[c] <= upper_range[i])\n",
    "            )\n",
    "        frac_hosp = float(num_hosp_in_range)/num_hosp*100.0\n",
    "        df_count.loc[i, c] = frac_hosp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "& \\multicolumn{5}{c}{Coverage} \\\\ \\hline\n",
      "Table Name &\n",
      "No data (0\\%) &\n",
      "Low coverage (0-20\\%) &\n",
      "Medium coverage (20-60\\%) &\n",
      "High coverage (60-80\\%) &\n",
      "Excellent coverage (80-100\\%) \\\\ \\hline\n",
      "\\tblname{admissiondx}                    &  0.48 &  0.48 &  5.77 & 15.38 & 77.88 \\\\\n",
      "\\tblname{admissiondrug}                  & 41.35 & 24.52 & 19.23 &  2.88 & 12.02 \\\\\n",
      "\\tblname{allergy}                        & 10.58 & 20.67 & 63.46 &  5.29 &  0.00 \\\\\n",
      "\\tblname{apacheapsvar}                   &  0.00 &  0.48 &  6.73 & 14.90 & 77.88 \\\\\n",
      "\\tblname{apachepredvar}                  &  0.00 &  0.48 &  6.73 & 14.90 & 77.88 \\\\\n",
      "\\tblname{apachepatientresult}            &  8.65 &  0.96 & 16.83 & 12.98 & 60.58 \\\\\n",
      "\\tblname{careplancareprovider}           &  0.96 &  0.96 & 12.02 & 12.98 & 73.08 \\\\\n",
      "\\tblname{careplaneol}                    & 53.85 & 46.15 &  0.00 &  0.00 &  0.00 \\\\\n",
      "\\tblname{careplangeneral}                &  0.48 &  0.00 &  0.48 &  2.40 & 96.63 \\\\\n",
      "\\tblname{careplangoal}                   & 62.98 & 27.40 &  0.96 &  4.33 &  4.33 \\\\\n",
      "\\tblname{careplaninfectiousdisease}      & 53.85 & 38.94 &  6.73 &  0.48 &  0.00 \\\\\n",
      "\\tblname{customlab}                      & 92.79 &  4.81 &  0.48 &  0.48 &  1.44 \\\\\n",
      "\\tblname{diagnosis}                      &  0.48 &  0.48 & 11.54 & 11.54 & 75.96 \\\\\n",
      "\\tblname{infusiondrug}                   & 26.92 & 16.35 & 40.38 &  9.62 &  6.73 \\\\\n",
      "\\tblname{intakeoutput}                   &  2.40 &  3.85 &  5.29 & 12.02 & 76.44 \\\\\n",
      "\\tblname{lab}                            &  0.48 &  0.00 &  0.48 &  2.88 & 96.15 \\\\\n",
      "\\tblname{medication}                     & 16.35 &  7.21 &  2.40 &  1.92 & 72.12 \\\\\n",
      "\\tblname{microlab}                       & 89.42 &  5.77 &  3.85 &  0.96 &  0.00 \\\\\n",
      "\\tblname{note}                           &  0.00 &  0.00 &  3.37 & 16.83 & 79.81 \\\\\n",
      "\\tblname{nurseassessment}                & 92.31 &  1.92 &  0.96 &  0.00 &  4.81 \\\\\n",
      "\\tblname{nursecare}                      & 93.27 &  0.96 &  0.96 &  0.00 &  4.81 \\\\\n",
      "\\tblname{nursecharting}                  &  0.48 &  0.96 &  1.92 &  4.33 & 92.31 \\\\\n",
      "\\tblname{pasthistory}                    &  0.48 &  0.48 &  4.33 & 17.31 & 77.40 \\\\\n",
      "\\tblname{physicalexam}                   &  0.48 &  0.48 &  3.85 & 17.79 & 77.40 \\\\\n",
      "\\tblname{respiratorycare}                & 24.52 & 41.35 & 33.17 &  0.48 &  0.48 \\\\\n",
      "\\tblname{respiratorycharting}            & 11.06 & 15.38 & 35.58 &  9.62 & 28.37 \\\\\n",
      "\\tblname{treatment}                      &  6.25 &  3.37 & 12.98 & 11.54 & 65.87 \\\\\n",
      "\\tblname{vitalaperiodic}                 &  0.96 &  0.00 &  3.85 &  5.29 & 89.90 \\\\\n",
      "\\tblname{vitalperiodic}                  &  0.96 &  0.00 &  3.37 &  2.40 & 93.27 \\\\\n"
     ]
    }
   ],
   "source": [
    "# print out header\n",
    "print('& \\\\multicolumn{5}{c}{Coverage} \\\\\\\\ \\\\hline')\n",
    "print('Table Name &')\n",
    "print('No data (0\\%) &')\n",
    "print('Low coverage (0-20\\%) &')\n",
    "print('Medium coverage (20-60\\%) &')\n",
    "print('High coverage (60-80\\%) &')\n",
    "print('Excellent coverage (80-100\\%) \\\\\\\\ \\\\hline')\n",
    "\n",
    "# print each row\n",
    "for i, row in df_count.T.iterrows():\n",
    "    print('{:40s}'.format( '\\\\tblname{' + row.name + '}'), end=' ')\n",
    "    for r in row:\n",
    "        print('& {:5.2f}'.format(r), end=' ')\n",
    "    print('\\\\\\\\')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>admissiondx</th>\n",
       "      <th>admissiondrug</th>\n",
       "      <th>allergy</th>\n",
       "      <th>apacheapsvar</th>\n",
       "      <th>apachepredvar</th>\n",
       "      <th>apachepatientresult</th>\n",
       "      <th>careplancareprovider</th>\n",
       "      <th>careplaneol</th>\n",
       "      <th>careplangeneral</th>\n",
       "      <th>careplangoal</th>\n",
       "      <th>...</th>\n",
       "      <th>nurseassessment</th>\n",
       "      <th>nursecare</th>\n",
       "      <th>nursecharting</th>\n",
       "      <th>pasthistory</th>\n",
       "      <th>physicalexam</th>\n",
       "      <th>respiratorycare</th>\n",
       "      <th>respiratorycharting</th>\n",
       "      <th>treatment</th>\n",
       "      <th>vitalaperiodic</th>\n",
       "      <th>vitalperiodic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>105</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>137</td>\n",
       "      <td>1</td>\n",
       "      <td>161</td>\n",
       "      <td>...</td>\n",
       "      <td>236</td>\n",
       "      <td>238</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>62</td>\n",
       "      <td>28</td>\n",
       "      <td>16</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>62</td>\n",
       "      <td>52</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>118</td>\n",
       "      <td>0</td>\n",
       "      <td>70</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>105</td>\n",
       "      <td>39</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14</td>\n",
       "      <td>49</td>\n",
       "      <td>162</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>43</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>11</td>\n",
       "      <td>9</td>\n",
       "      <td>84</td>\n",
       "      <td>91</td>\n",
       "      <td>33</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>39</td>\n",
       "      <td>7</td>\n",
       "      <td>13</td>\n",
       "      <td>38</td>\n",
       "      <td>38</td>\n",
       "      <td>33</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>44</td>\n",
       "      <td>45</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>29</td>\n",
       "      <td>13</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>199</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>199</td>\n",
       "      <td>199</td>\n",
       "      <td>155</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>247</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>236</td>\n",
       "      <td>198</td>\n",
       "      <td>198</td>\n",
       "      <td>1</td>\n",
       "      <td>72</td>\n",
       "      <td>168</td>\n",
       "      <td>230</td>\n",
       "      <td>238</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   admissiondx  admissiondrug  allergy  apacheapsvar  apachepredvar  \\\n",
       "0            1            105       27             0              0   \n",
       "1            1             62       52             1              1   \n",
       "2           14             49      162            17             17   \n",
       "3           39              7       13            38             38   \n",
       "4          199             30        0           199            199   \n",
       "\n",
       "   apachepatientresult  careplancareprovider  careplaneol  careplangeneral  \\\n",
       "0                   22                     2          137                1   \n",
       "1                    2                     2          118                0   \n",
       "2                   43                    30            0                1   \n",
       "3                   33                    33            0                6   \n",
       "4                  155                   187            0              247   \n",
       "\n",
       "   careplangoal      ...        nurseassessment  nursecare  nursecharting  \\\n",
       "0           161      ...                    236        238              1   \n",
       "1            70      ...                      4          2              2   \n",
       "2             2      ...                      2          2              4   \n",
       "3            11      ...                      0          0             11   \n",
       "4            11      ...                     12         12            236   \n",
       "\n",
       "   pasthistory  physicalexam  respiratorycare  respiratorycharting  treatment  \\\n",
       "0            1             1               62                   28         16   \n",
       "1            1             1              105                   39          8   \n",
       "2           11             9               84                   91         33   \n",
       "3           44            45                1                   24         29   \n",
       "4          198           198                1                   72        168   \n",
       "\n",
       "   vitalaperiodic  vitalperiodic  \n",
       "0               2              2  \n",
       "1               0              0  \n",
       "2               9              8  \n",
       "3              13              6  \n",
       "4             230            238  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_heatmap = df_count.copy()\n",
    "df_heatmap = (df_heatmap*256/100).astype(int)\n",
    "df_heatmap\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[12,8])\n",
    "plt.plot()\n",
    "plt.imshow(df_heatmap.T, interpolation='none', aspect='auto')\n",
    "plt.yticks(np.arange(df_heatmap.shape[1]), df_heatmap.columns)\n",
    "plt.xticks(np.arange(5), ['0%', '0-20%','20-60%','60-80%','80-100%'])\n",
    "# disable grid\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Figure for a single patient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# select a single ICU stay\n",
    "patientunitstayid = 142493\n",
    "query_schema = 'set search_path to eicu_crd_demo;'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Diagnosis: Complications of prev. peripheral vasc. surgery,surgery for (i.e.ligation of bleeder, exploration and evacuation of hematoma, debridement, pseudoaneurysms, clots, fistula, etc.)\n"
     ]
    }
   ],
   "source": [
    "# print APACHE diagnosis for patient\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", apacheadmissiondx as dx\n",
    "FROM patient\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "\"\"\"\n",
    "\n",
    "dx = pd.read_sql_query(query,con)\n",
    "print('Diagnosis: {}'.format(dx.loc[0,'dx']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>label</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>15</td>\n",
       "      <td>bedside glucose</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>50</td>\n",
       "      <td>Hgb</td>\n",
       "      <td>8.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>50</td>\n",
       "      <td>Hct</td>\n",
       "      <td>23.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>50</td>\n",
       "      <td>platelets x 1000</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>180</td>\n",
       "      <td>paCO2</td>\n",
       "      <td>28.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset             label  value\n",
       "0             142493      15   bedside glucose  145.0\n",
       "1             142493      50               Hgb    8.2\n",
       "2             142493      50               Hct   23.7\n",
       "3             142493      50  platelets x 1000   65.0\n",
       "4             142493     180             paCO2   28.0"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull lab data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", labresultoffset as \"offset\"\n",
    ", labname as \"label\"\n",
    ", labresult as \"value\"\n",
    "FROM lab\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "lab = pd.read_sql_query(query,con)\n",
    "lab.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>label</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>115</td>\n",
       "      <td>3 ML  -  IPRATROPIUM-ALBUTEROL 0.5-2.5 (3) MG/...</td>\n",
       "      <td>3 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>312</td>\n",
       "      <td>MORPHINE INJ</td>\n",
       "      <td>3 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>668</td>\n",
       "      <td>1 ML  -  NALOXONE HCL 0.4 MG/ML IJ SOLN</td>\n",
       "      <td>0.1 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>670</td>\n",
       "      <td>OXYCODONE HCL 5 MG PO TABS</td>\n",
       "      <td>5 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>1122</td>\n",
       "      <td>CLOPIDOGREL BISULFATE 75 MG PO TABS</td>\n",
       "      <td>75 3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset  \\\n",
       "0             142493     115   \n",
       "1             142493     312   \n",
       "2             142493     668   \n",
       "3             142493     670   \n",
       "4             142493    1122   \n",
       "\n",
       "                                               label  value  \n",
       "0  3 ML  -  IPRATROPIUM-ALBUTEROL 0.5-2.5 (3) MG/...    3 1  \n",
       "1                                       MORPHINE INJ    3 3  \n",
       "2            1 ML  -  NALOXONE HCL 0.4 MG/ML IJ SOLN  0.1 3  \n",
       "3                         OXYCODONE HCL 5 MG PO TABS    5 3  \n",
       "4                CLOPIDOGREL BISULFATE 75 MG PO TABS   75 3  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull meds data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", drugstartoffset as \"offset\"\n",
    ", drugname as \"label\"\n",
    ", dosage as \"value\"\n",
    "FROM medication\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "AND drugname IS NOT NULL\n",
    "AND dosage IS NOT NULL\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "med = pd.read_sql_query(query,con)\n",
    "med.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>label</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>87</td>\n",
       "      <td>Volume-Transfuse platelet pheresis</td>\n",
       "      <td>250.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>132</td>\n",
       "      <td>Volume-Transfuse platelet pheresis</td>\n",
       "      <td>250.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>237</td>\n",
       "      <td>Volume-Transfuse red blood cells</td>\n",
       "      <td>350.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>342</td>\n",
       "      <td>P.O.</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>582</td>\n",
       "      <td>P.O.</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset                               label  value\n",
       "0             142493      87  Volume-Transfuse platelet pheresis  250.0\n",
       "1             142493     132  Volume-Transfuse platelet pheresis  250.0\n",
       "2             142493     237    Volume-Transfuse red blood cells  350.0\n",
       "3             142493     342                                P.O.   50.0\n",
       "4             142493     582                                P.O.   50.0"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull meds data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", intakeoutputoffset as \"offset\"\n",
    ", celllabel as \"label\"\n",
    ", cellvaluenumeric as \"value\"\n",
    "FROM intakeoutput\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "AND celllabel IS NOT NULL\n",
    "AND cellvaluenumeric IS NOT NULL\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "io = pd.read_sql_query(query,con)\n",
    "io.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>label</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [patientunitstayid, offset, label, value]\n",
       "Index: []"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull infusion drug data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", infusionoffset as \"offset\"\n",
    ", drugname as \"label\"\n",
    ", coalesce(drugrate,infusionrate) as \"value\"\n",
    "FROM infusiondrug\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "AND coalesce(drugrate, infusionrate) IS NOT NULL\n",
    "AND drugname IS NOT NULL\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "inf = pd.read_sql_query(query,con)\n",
    "inf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>sysbp</th>\n",
       "      <th>diasbp</th>\n",
       "      <th>meanbp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>116</td>\n",
       "      <td>121.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>242</td>\n",
       "      <td>120.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>321</td>\n",
       "      <td>110.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>79.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>352</td>\n",
       "      <td>118.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>462</td>\n",
       "      <td>130.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>88.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset  sysbp  diasbp  meanbp\n",
       "0             142493     116  121.0    58.0    83.0\n",
       "1             142493     242  120.0    58.0    83.0\n",
       "2             142493     321  110.0    55.0    79.0\n",
       "3             142493     352  118.0    58.0    83.0\n",
       "4             142493     462  130.0    61.0    88.0"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull vitalaperiodic data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", observationoffset as \"offset\"\n",
    ", noninvasivesystolic sysbp\n",
    ", noninvasivediastolic diasbp\n",
    ", noninvasivemean meanbp\n",
    "FROM vitalaperiodic\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "AND noninvasivemean IS NOT NULL\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "vap = pd.read_sql_query(query,con)\n",
    "vap.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>sysbp</th>\n",
       "      <th>diasbp</th>\n",
       "      <th>meanbp</th>\n",
       "      <th>heartrate</th>\n",
       "      <th>spo2</th>\n",
       "      <th>respiration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>5</td>\n",
       "      <td>135.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>83</td>\n",
       "      <td>100.0</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>10</td>\n",
       "      <td>159.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>91</td>\n",
       "      <td>100.0</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>15</td>\n",
       "      <td>169.0</td>\n",
       "      <td>74.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>97</td>\n",
       "      <td>100.0</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>20</td>\n",
       "      <td>134.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>90</td>\n",
       "      <td>98.0</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>25</td>\n",
       "      <td>125.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>96</td>\n",
       "      <td>98.0</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset  sysbp  diasbp  meanbp  heartrate   spo2  \\\n",
       "0             142493       5  135.0    50.0    83.0         83  100.0   \n",
       "1             142493      10  159.0    70.0   105.0         91  100.0   \n",
       "2             142493      15  169.0    74.0   114.0         97  100.0   \n",
       "3             142493      20  134.0    56.0    88.0         90   98.0   \n",
       "4             142493      25  125.0    55.0    83.0         96   98.0   \n",
       "\n",
       "   respiration  \n",
       "0           25  \n",
       "1           20  \n",
       "2           27  \n",
       "3           25  \n",
       "4           27  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pull vitalperiodic data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", observationoffset as \"offset\"\n",
    ", systemicsystolic sysbp\n",
    ", systemicdiastolic diasbp\n",
    ", systemicmean meanbp\n",
    ", heartrate\n",
    ", sao2 as spo2\n",
    ", respiration\n",
    "FROM vitalperiodic\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "vp = pd.read_sql_query(query,con)\n",
    "vp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>patientunitstayid</th>\n",
       "      <th>offset</th>\n",
       "      <th>heartrate</th>\n",
       "      <th>respiratoryrate</th>\n",
       "      <th>invasivebpsystolic</th>\n",
       "      <th>invasivebpdiastolic</th>\n",
       "      <th>invasivebpmean</th>\n",
       "      <th>spo2</th>\n",
       "      <th>noninvasivebpdiastolic</th>\n",
       "      <th>noninvasivebpsystolic</th>\n",
       "      <th>noninvasivebpmean</th>\n",
       "      <th>temperature</th>\n",
       "      <th>paingoal</th>\n",
       "      <th>painscore</th>\n",
       "      <th>verbal</th>\n",
       "      <th>eyes</th>\n",
       "      <th>gcs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>142493</td>\n",
       "      <td>-8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>142493</td>\n",
       "      <td>-3</td>\n",
       "      <td>66.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>132.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>142493</td>\n",
       "      <td>2</td>\n",
       "      <td>83.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>142493</td>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>87.0</td>\n",
       "      <td>36.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>142493</td>\n",
       "      <td>8</td>\n",
       "      <td>99.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>77.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36.6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   patientunitstayid  offset  heartrate  respiratoryrate  invasivebpsystolic  \\\n",
       "0             142493      -8        NaN              NaN                 NaN   \n",
       "1             142493      -3       66.0              NaN               132.0   \n",
       "2             142493       2       83.0             25.0               130.0   \n",
       "3             142493       7        NaN             16.0                 NaN   \n",
       "4             142493       8       99.0             17.0                 NaN   \n",
       "\n",
       "   invasivebpdiastolic  invasivebpmean   spo2  noninvasivebpdiastolic  \\\n",
       "0                  NaN             NaN  100.0                     NaN   \n",
       "1                 54.0            79.0  100.0                     NaN   \n",
       "2                 49.0            79.0  100.0                     NaN   \n",
       "3                  NaN             NaN    NaN                     NaN   \n",
       "4                  NaN             NaN    NaN                    77.0   \n",
       "\n",
       "   noninvasivebpsystolic  noninvasivebpmean  temperature  paingoal  painscore  \\\n",
       "0                    NaN                NaN          NaN       NaN        NaN   \n",
       "1                    NaN                NaN          NaN       NaN        NaN   \n",
       "2                    NaN                NaN          NaN       NaN        NaN   \n",
       "3                    NaN               87.0         36.8       NaN        NaN   \n",
       "4                  170.0                NaN         36.6       NaN        NaN   \n",
       "\n",
       "   verbal  eyes  gcs  \n",
       "0     NaN   NaN  NaN  \n",
       "1     NaN   NaN  NaN  \n",
       "2     NaN   NaN  NaN  \n",
       "3     NaN   NaN  NaN  \n",
       "4     NaN   NaN  NaN  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# nursecharting data\n",
    "query = query_schema + \\\n",
    "\"\"\"\n",
    "SELECT patientunitstayid\n",
    ", nursingchartoffset as \"offset\"\n",
    ", avg((case when nursingchartcelltypevallabel = 'Heart Rate' AND nursingchartcelltypevalname = 'Heart Rate' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as HeartRate                 \n",
    ", avg((case when nursingchartcelltypevallabel = 'Respiratory Rate' AND nursingchartcelltypevalname = 'Respiratory Rate' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as RespiratoryRate           \n",
    ", avg((case when nursingchartcelltypevallabel = 'Invasive BP' AND nursingchartcelltypevalname = 'Invasive BP Systolic' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as InvasiveBPSystolic       \n",
    ", avg((case when nursingchartcelltypevallabel = 'Invasive BP' AND nursingchartcelltypevalname = 'Invasive BP Diastolic' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as InvasiveBPDiastolic       \n",
    ", avg((case when nursingchartcelltypevallabel = 'Arterial Line MAP (mmHg)' AND nursingchartcelltypevalname = 'Value' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as InvasiveBPMean    \n",
    ", avg((case when nursingchartcelltypevallabel = 'O2 Saturation' AND nursingchartcelltypevalname = 'O2 Saturation' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as spo2\n",
    ", avg((case when nursingchartcelltypevallabel = 'Non-Invasive BP' AND nursingchartcelltypevalname = 'Non-Invasive BP Diastolic' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as NonInvasiveBPDiastolic  \n",
    ", avg((case when nursingchartcelltypevallabel = 'Non-Invasive BP' AND nursingchartcelltypevalname = 'Non-Invasive BP Systolic' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as NonInvasiveBPSystolic\n",
    ", avg((case when nursingchartcelltypevallabel = 'Non-Invasive BP' AND nursingchartcelltypevalname = 'Non-Invasive BP Mean' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as NonInvasiveBPMean\n",
    ", avg((case when nursingchartcelltypevallabel = 'Temperature' AND nursingchartcelltypevalname = 'Temperature (C)' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as Temperature\n",
    ", avg((case when nursingchartcelltypevallabel = 'Patient s Comfort/Function (Pain) GOAL At Rest' AND nursingchartcelltypevalname = 'Value' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as paingoal                      \n",
    ", avg((case when nursingchartcelltypevallabel = 'Pain Score/Goal' AND nursingchartcelltypevalname = 'Pain Score' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as PainScore\n",
    ", avg((case when nursingchartcelltypevallabel = 'Verbal Response' AND nursingchartcelltypevalname = 'Value' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as verbal\n",
    ", avg((case when nursingchartcelltypevallabel = 'Eye Opening' AND nursingchartcelltypevalname = 'Value' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as eyes\n",
    ", avg((case when nursingchartcelltypevallabel = 'Glasgow coma score' AND nursingchartcelltypevalname = 'GCS Total' THEN nursingchartvalue ELSE NULL END)::NUMERIC) as gcs\n",
    "FROM nursecharting\n",
    "WHERE patientunitstayid = \"\"\" + str(patientunitstayid) + \"\"\"\n",
    "GROUP BY patientunitstayid, nursingchartoffset\n",
    "ORDER BY \"offset\"\n",
    "\"\"\"\n",
    "\n",
    "nc = pd.read_sql_query(query,con)\n",
    "nc.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the data\n",
    "\n",
    "# convert from minutes to our time unit here\n",
    "TIME_FACTOR = 1.0/60.0\n",
    "TIME_UNIT = 'hours'\n",
    "\n",
    "plt.figure(figsize=[12,10])\n",
    "p_hr = plt.plot(vp['offset']*TIME_FACTOR, vp['heartrate'], '.--', label='_no_legend_')\n",
    "p_nibp = plt.plot(vap['offset']*TIME_FACTOR, vap['meanbp'], '.--', label='_no_legend_')\n",
    "p_ibp = plt.plot(vp['offset']*TIME_FACTOR, vp['meanbp'], '.--', label='_no_legend_')\n",
    "p_o2 = plt.plot(vp['offset']*TIME_FACTOR, vp['spo2'], '.--', label='_no_legend_')\n",
    "p_rr = plt.plot(vp['offset']*TIME_FACTOR, vp['respiration'], '.--', label='_no_legend_')\n",
    "\n",
    "# nurse validated\n",
    "plt.plot(nc['offset']*TIME_FACTOR, nc['heartrate'], '^',\n",
    "         markersize=12, markerfacecolor=None, color=p_hr[0].get_color(),\n",
    "         label='Heart rate, beats/minute')\n",
    "plt.plot(nc['offset']*TIME_FACTOR, nc['invasivebpmean'], 's',\n",
    "         markersize=12, markerfacecolor=None, color=p_ibp[0].get_color(), label='Mean blood pressure (non-invasive), mmHg')\n",
    "plt.plot(nc['offset']*TIME_FACTOR, nc['noninvasivebpmean'], 'o',\n",
    "         markersize=12, markerfacecolor=None, color=p_nibp[0].get_color(), label='Mean blood pressure (invasive), mmHg')\n",
    "plt.plot(nc['offset']*TIME_FACTOR, nc['spo2'], 'd',\n",
    "         markersize=12, markerfacecolor=None, color=p_o2[0].get_color(), label='Oxygen saturation, %')\n",
    "plt.plot(nc['offset']*TIME_FACTOR, nc['respiratoryrate'], 'v',\n",
    "         markersize=12, markerfacecolor=None, color=p_rr[0].get_color(), label='Respiration rate, breaths/min')\n",
    "\"\"\"\n",
    "# Plot infusions as a scaled line\n",
    "infdrg_dict = OrderedDict([\n",
    "        ['Norepinephrine, mcg/min', [145, 'Norepinephrine (mcg/min)', 0, 10 ]]\n",
    "        #['Normal Saline (ml/hr)', [155, 'NS (ml/hr)', 0, 500]]\n",
    "    ])\n",
    "\n",
    "xloc = 0\n",
    "for item in infdrg_dict:\n",
    "    yloc = infdrg_dict[item][0]\n",
    "    label = infdrg_dict[item][1]\n",
    "    plot_lo = infdrg_dict[item][2]\n",
    "    plot_hi = infdrg_dict[item][3]\n",
    "    \n",
    "    scale_data = 10.0/(plot_hi-plot_lo)\n",
    "    plt.text(xloc, yloc, item, fontsize=18, horizontalalignment='right')\n",
    "    # plot the trend over time\n",
    "    df_item = inf.loc[inf['label']==label,:]\n",
    "    # limit to first 2 days\n",
    "    df_item = df_item.loc[df_item['offset']<=(1440*2),:]\n",
    "    scaled_data = df_item['value'].astype(float)\n",
    "    # shift low to 0\n",
    "    scaled_data = scaled_data - plot_lo\n",
    "    # shift high to 1\n",
    "    scaled_data = scaled_data / plot_hi\n",
    "    # move up to [yloc, yloc+5]\n",
    "    scaled_data = (scaled_data)*5.0 + yloc\n",
    "    \n",
    "    plt.plot( df_item['offset']*TIME_FACTOR, scaled_data, 'o--', label='_nolegend_' )\n",
    "    \n",
    "# Plot intake output as a single marker indicating start time\n",
    "med_dict = OrderedDict([\n",
    "        ['Normal Saline (ml/hr)', [155, 'NS (ml/hr)', 0, 500]]\n",
    "    ])\n",
    "\n",
    "# plot a single marker for each instance - regardless of the value of the instance\n",
    "xloc = 0\n",
    "for item in med_dict:\n",
    "    yloc = med_dict[item][0]\n",
    "    label = med_dict[item][1]\n",
    "    plt.text(xloc, yloc, item, fontsize=18, horizontalalignment='right')\n",
    "    \n",
    "    df_item = med.loc[med['label']==label,:]\n",
    "    \n",
    "    # limit to first 2 days\n",
    "    df_item = df_item.loc[(df_item['offset']>0) & (df_item['offset']<=(1440*2)),:]\n",
    "    \n",
    "    plt.plot( df_item['offset']*TIME_FACTOR, (yloc+2.0)*np.ones(df_item.shape[0]), 'o',\n",
    "             label='_nolegend_' )\n",
    "\"\"\"\n",
    "\n",
    "# Plot intake output as a single marker indicating start time\n",
    "io_dict = OrderedDict([\n",
    "        ['Red blood cell transfusion', [125, 'Volume-Transfuse red blood cells']]\n",
    "        #['Normal Saline (ml/hr)', [155, 'NS (ml/hr)', 0, 500]]\n",
    "    ])\n",
    "\n",
    "# plot a single marker for each instance - regardless of the value of the instance\n",
    "xloc = 0\n",
    "for item in io_dict:\n",
    "    yloc = io_dict[item][0]\n",
    "    label = io_dict[item][1]\n",
    "    plt.text(xloc, yloc, item, fontsize=18, horizontalalignment='right', color='#555555')\n",
    "    \n",
    "    df_item = io.loc[io['label']==label,:]\n",
    "    \n",
    "    # limit to first 2 days\n",
    "    df_item = df_item.loc[(df_item['offset']>0) & (df_item['offset']<=(1440*2)),:]\n",
    "    \n",
    "    plt.plot( df_item['offset']*TIME_FACTOR, (yloc+2.0)*np.ones(df_item.shape[0]), '+',\n",
    "             markersize=20, mew=8,\n",
    "             label='_nolegend_' )\n",
    "\n",
    "# Plot labs as text on the plot\n",
    "labs = OrderedDict([\n",
    "        #['Creatinine, mg/dL', [195, 'creatinine']],\n",
    "        #['White blood cell, K/uL', [ 185, 'WBC x 1000' ]],\n",
    "        #['Troponin, g/dL', [175, 'troponin - I']],\n",
    "        #['Lactate, mmol/L', [175, 'lactate']],\n",
    "        ['PaO2, mmHg', [155, 'paO2']],\n",
    "        ['Hemoglobin, g/dL', [145, 'Hgb']],\n",
    "        ['Hematocrit, %', [135, 'Hct']]\n",
    "    ])\n",
    "\n",
    "xloc = 0\n",
    "for l in labs:\n",
    "    yloc = labs[l][0]\n",
    "    label = labs[l][1]\n",
    "    \n",
    "    # plot the trend over time\n",
    "    df_item = lab.loc[lab['label']==label,:]\n",
    "    # limit to first 2 days\n",
    "    df_item = df_item.loc[(df_item['offset']<=(1440*2)) & (df_item['offset']>0),:]\n",
    "    \n",
    "    plt.text(xloc, yloc, l, fontsize=18, horizontalalignment='right', color='#555555')\n",
    "    for i, row in df_item.iterrows():\n",
    "        # put text for pao2 without a decimal\n",
    "        if l=='PaO2, mmHg':\n",
    "            v = int(row['value'])\n",
    "        else:\n",
    "            v = row['value']\n",
    "            \n",
    "        plt.text(row['offset']*TIME_FACTOR, yloc, v,\n",
    "                 fontsize=16, color='#555555', label='_nolegend_') \n",
    "\n",
    "plt.yticks(np.arange(0, 120, 20))\n",
    "plt.xticks(np.arange(0, 12.01, 4))\n",
    "plt.xlim([0,12])\n",
    "plt.ylim([0,170])\n",
    "plt.xlabel('Time since unit admission (' + TIME_UNIT + ')')\n",
    "plt.ylabel('Measurement value')\n",
    "# manually position the y-label\n",
    "ax = plt.gca()\n",
    "ax.yaxis.set_label_coords(-0.1, 0.3)\n",
    "leg = plt.legend(loc='lower right')\n",
    "plt.setp(leg.get_texts(), color='#555555')\n",
    "plt.savefig('ExamplePatient.pdf', bbox_inches='tight')\n",
    "plt.savefig('ExamplePatient.png', bbox_inches='tight')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}