plot_meds.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a12b30cb-0077-4515-a827-4a526c810431",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from tqdm.notebook import tqdm\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0cbd9f8e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Antiarrhythmics': ['Amiodarone 600/500',\n",
       "  'Amiodarone',\n",
       "  'Lidocaine',\n",
       "  'Adenosine',\n",
       "  'Procainamide'],\n",
       " 'Electrolyte Supplements': ['Potassium Chloride',\n",
       "  'Calcium Gluconate',\n",
       "  'Calcium Gluconate (CRRT)',\n",
       "  'K Phos',\n",
       "  'Na Phos'],\n",
       " 'Beta-Blockers': ['Metoprolol', 'Labetalol', 'Esmolol'],\n",
       " 'Diuretics': ['Furosemide (Lasix)', 'Furosemide (Lasix) 250/50', 'Mannitol'],\n",
       " 'Cardiac Glycosides': ['Digoxin (Lanoxin)'],\n",
       " 'Insulins': ['Insulin - Regular',\n",
       "  'Insulin - Humalog',\n",
       "  'Insulin - Glargine',\n",
       "  'Insulin - NPH',\n",
       "  'Insulin - 70/30',\n",
       "  'Insulin - Novolog',\n",
       "  'Insulin - Humalog 75/25'],\n",
       " 'Opioid Analgesics': ['Fentanyl',\n",
       "  'Fentanyl (Concentrate)',\n",
       "  'Hydromorphone (Dilaudid)',\n",
       "  'Morphine Sulfate',\n",
       "  'Meperidine (Demerol)',\n",
       "  'Methadone Hydrochloride'],\n",
       " 'Sedatives and Anesthetics': ['Propofol',\n",
       "  'Midazolam (Versed)',\n",
       "  'Dexmedetomidine (Precedex)',\n",
       "  'Lorazepam (Ativan)',\n",
       "  'Ketamine',\n",
       "  'Diazepam (Valium)',\n",
       "  'Pentobarbital'],\n",
       " 'Thrombolytics': ['Alteplase (TPA)'],\n",
       " 'Vasopressors and Inotropes': ['Norepinephrine',\n",
       "  'Phenylephrine',\n",
       "  'Epinephrine',\n",
       "  'Dopamine',\n",
       "  'Vasopressin',\n",
       "  'Milrinone',\n",
       "  'Dobutamine',\n",
       "  'Isuprel']}"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import json\n",
    "\n",
    "with open('med_categories.json','r') as f:\n",
    "    med_categories = json.load(f)\n",
    "med_categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c4613b66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x1000 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,axs = plt.subplots(3, 4, figsize=(14,10), gridspec_kw={'wspace': 0, 'hspace': 0})\n",
    "cp = sns.color_palette()\n",
    "\n",
    "for (parent_category, subcategories), ax in zip(med_categories.items(), axs.flatten()):\n",
    "    ax.set_ylim([0,50])\n",
    "    ax.set_xlim([0,50])\n",
    "    ax.text(5,45,parent_category,fontsize=12,fontweight=500)\n",
    "    for i, cat in enumerate(subcategories):\n",
    "        ax.text(10,40-(5*i),cat,fontsize=11,color=cp[i])\n",
    "\n",
    "for ax in axs.flatten():\n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "\n",
    "fig.tight_layout()\n",
    "# fig.savefig('images/med_names.pdf',bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "731a0f49",
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_datetime(date_str, time_str):\n",
    "    try:\n",
    "        date = pd.to_datetime(date_str, format='%Y%m%d')\n",
    "        hours = int(time_str // 10000)\n",
    "        minutes = int((time_str % 10000) // 100)\n",
    "        seconds = int(time_str % 100)\n",
    "        time = pd.to_timedelta(f'{hours}:{minutes}:{seconds}')\n",
    "        return date + time\n",
    "    except Exception as e:\n",
    "        print(f\"Fehler beim Parsing: {e}\")\n",
    "        return pd.NaT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a95a10ba-09be-4f6c-b1d3-f3baa02082c0",
   "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>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>stay_id</th>\n",
       "      <th>id_ehr</th>\n",
       "      <th>id_cxr</th>\n",
       "      <th>last_careunit</th>\n",
       "      <th>intime</th>\n",
       "      <th>outtime</th>\n",
       "      <th>los</th>\n",
       "      <th>admittime</th>\n",
       "      <th>...</th>\n",
       "      <th>last_echo_id</th>\n",
       "      <th>last_ecg_id</th>\n",
       "      <th>split</th>\n",
       "      <th>study_id_cxr</th>\n",
       "      <th>demographics_text2</th>\n",
       "      <th>demographics_text</th>\n",
       "      <th>cxr_text</th>\n",
       "      <th>ecg_text</th>\n",
       "      <th>med_text</th>\n",
       "      <th>icd_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18730522</td>\n",
       "      <td>21463350</td>\n",
       "      <td>30004391</td>\n",
       "      <td>18730522_episode2_timeseries.csv</td>\n",
       "      <td>d2bae3a3-3917d71b-f44edcd7-81c7017a-f15288e7</td>\n",
       "      <td>Medical Intensive Care Unit (MICU)</td>\n",
       "      <td>2153-09-05 13:12:00</td>\n",
       "      <td>2153-09-13 18:21:18</td>\n",
       "      <td>8.214792</td>\n",
       "      <td>2153-09-05 11:58:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>train</td>\n",
       "      <td>51967233</td>\n",
       "      <td>A Patient was admitted to the ICU in the Medic...</td>\n",
       "      <td>A male, white, 89 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: Left PICC...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>The patient received Dexmedetomidine (Precedex...</td>\n",
       "      <td>The patient's diagnoses are Sepsis due to Esch...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12509799</td>\n",
       "      <td>25897223</td>\n",
       "      <td>30004530</td>\n",
       "      <td>12509799_episode1_timeseries.csv</td>\n",
       "      <td>b07e6c45-cab5a33c-bb1cade8-a043a627-48572ac5</td>\n",
       "      <td>Cardiac Vascular Intensive Care Unit (CVICU)</td>\n",
       "      <td>2165-07-31 09:40:35</td>\n",
       "      <td>2165-08-03 16:29:09</td>\n",
       "      <td>3.283727</td>\n",
       "      <td>2165-07-29 06:44:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43916976.0</td>\n",
       "      <td>train</td>\n",
       "      <td>55465259</td>\n",
       "      <td>A Patient was admitted to the ICU in the Cardi...</td>\n",
       "      <td>A male, white, 63 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: None.</td>\n",
       "      <td>The ECG shows a Sinus rhythm, Left bundle bran...</td>\n",
       "      <td>The patient received Acetaminophen-IV, Calcium...</td>\n",
       "      <td>The patient's diagnoses are Coronary atheroscl...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10332722</td>\n",
       "      <td>29393377</td>\n",
       "      <td>30005362</td>\n",
       "      <td>10332722_episode1_timeseries.csv</td>\n",
       "      <td>e4e55bf2-3b86fa59-5138badb-62e90dd5-489e3902</td>\n",
       "      <td>Medical Intensive Care Unit (MICU)</td>\n",
       "      <td>2156-12-28 21:42:10</td>\n",
       "      <td>2157-01-03 19:16:43</td>\n",
       "      <td>5.898993</td>\n",
       "      <td>2156-12-15 23:03:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>40544966.0</td>\n",
       "      <td>train</td>\n",
       "      <td>57947776</td>\n",
       "      <td>A Patient was admitted to the ICU in the Medic...</td>\n",
       "      <td>A male, white, 80 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: None.</td>\n",
       "      <td>The ECG shows a Sinus tachycardia with PAC(s),...</td>\n",
       "      <td>The patient received Heparin Sodium, Heparin S...</td>\n",
       "      <td>The patient's diagnoses are Urinary tract infe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18855147</td>\n",
       "      <td>25702658</td>\n",
       "      <td>30005366</td>\n",
       "      <td>18855147_episode2_timeseries.csv</td>\n",
       "      <td>b361a1e1-b9c3ab9d-c2cc8344-2903cfd8-3888d7b9</td>\n",
       "      <td>Coronary Care Unit (CCU)</td>\n",
       "      <td>2202-12-27 17:36:59</td>\n",
       "      <td>2202-12-31 14:30:31</td>\n",
       "      <td>3.870509</td>\n",
       "      <td>2202-12-27 17:36:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>41783871.0</td>\n",
       "      <td>train</td>\n",
       "      <td>54043642</td>\n",
       "      <td>A Patient was admitted to the ICU in the Coron...</td>\n",
       "      <td>A female, white, 69 years old patient with a w...</td>\n",
       "      <td>CXR findings: None. CXR impressions: None.</td>\n",
       "      <td>The ECG shows a Sinus rhythm, Possible septal ...</td>\n",
       "      <td>The patient received Haloperidol (Haldol), Met...</td>\n",
       "      <td>The patient's diagnoses are Acute myocardial i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15524760</td>\n",
       "      <td>28718343</td>\n",
       "      <td>30006565</td>\n",
       "      <td>15524760_episode1_timeseries.csv</td>\n",
       "      <td>65dfa87f-d808fe8b-e2c80e3c-f66ae4ff-ea264e9c</td>\n",
       "      <td>Medical Intensive Care Unit (MICU)</td>\n",
       "      <td>2127-07-18 12:44:45</td>\n",
       "      <td>2127-07-21 20:32:57</td>\n",
       "      <td>3.325139</td>\n",
       "      <td>2127-07-17 22:53:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>train</td>\n",
       "      <td>52141331</td>\n",
       "      <td>A Patient was admitted to the ICU in the Medic...</td>\n",
       "      <td>A male, white, 53 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: Mild card...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>The patient received Famotidine (Pepcid), Fent...</td>\n",
       "      <td>The patient's diagnoses are Pneumonitis due to...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6793</th>\n",
       "      <td>12127320</td>\n",
       "      <td>23231713</td>\n",
       "      <td>39970416</td>\n",
       "      <td>12127320_episode1_timeseries.csv</td>\n",
       "      <td>9c8f4dc5-808f29b9-f340457c-52c4ad5e-c2ad0ce4</td>\n",
       "      <td>Surgical Intensive Care Unit (SICU)</td>\n",
       "      <td>2153-11-12 12:45:31</td>\n",
       "      <td>2153-11-15 20:41:11</td>\n",
       "      <td>3.330324</td>\n",
       "      <td>2153-11-12 06:54:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43475719.0</td>\n",
       "      <td>test</td>\n",
       "      <td>58201427</td>\n",
       "      <td>A Patient was admitted to the ICU in the Surgi...</td>\n",
       "      <td>A male, unable to obtain, 72 years old patient...</td>\n",
       "      <td>CXR findings: None. CXR impressions: None.</td>\n",
       "      <td>The ECG shows a Atrial fibrillation with rapid...</td>\n",
       "      <td>The patient received Famotidine (Pepcid), Halo...</td>\n",
       "      <td>The patient's diagnoses are Hypertensive encep...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6794</th>\n",
       "      <td>16438315</td>\n",
       "      <td>21184320</td>\n",
       "      <td>39972274</td>\n",
       "      <td>16438315_episode1_timeseries.csv</td>\n",
       "      <td>d91fae99-1f7929e2-d200297e-5569d6c9-d8c345e4</td>\n",
       "      <td>Trauma SICU (TSICU)</td>\n",
       "      <td>2185-03-09 01:32:00</td>\n",
       "      <td>2185-03-17 18:57:06</td>\n",
       "      <td>8.725764</td>\n",
       "      <td>2185-03-09 00:49:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>test</td>\n",
       "      <td>52834029</td>\n",
       "      <td>A Patient was admitted to the ICU in the Traum...</td>\n",
       "      <td>A female, black/african american, 33 years old...</td>\n",
       "      <td>CXR findings: The endotracheal tube ends 5 cm ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>The patient received Famotidine (Pepcid), Fent...</td>\n",
       "      <td>The patient's diagnoses are Cerebral cysts, No...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6795</th>\n",
       "      <td>14851188</td>\n",
       "      <td>22655854</td>\n",
       "      <td>39972327</td>\n",
       "      <td>14851188_episode1_timeseries.csv</td>\n",
       "      <td>c162887f-d90031ff-a5dae741-f20e7f14-b5022765</td>\n",
       "      <td>Cardiac Vascular Intensive Care Unit (CVICU)</td>\n",
       "      <td>2158-07-26 17:45:10</td>\n",
       "      <td>2158-07-31 20:10:25</td>\n",
       "      <td>5.100868</td>\n",
       "      <td>2158-07-26 17:44:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>41452079.0</td>\n",
       "      <td>test</td>\n",
       "      <td>54985656</td>\n",
       "      <td>A Patient was admitted to the ICU in the Cardi...</td>\n",
       "      <td>A male, white, 62 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: In compar...</td>\n",
       "      <td>The ECG shows a Sinus rhythm., Possible inferi...</td>\n",
       "      <td>The patient received Acetaminophen-IV, Calcium...</td>\n",
       "      <td>The patient's diagnoses are Abdominal aneurysm...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6796</th>\n",
       "      <td>18315784</td>\n",
       "      <td>28656232</td>\n",
       "      <td>39982305</td>\n",
       "      <td>18315784_episode1_timeseries.csv</td>\n",
       "      <td>4c373404-894fc136-26d5fb93-fe4f8dae-8d5cafbd</td>\n",
       "      <td>Surgical Intensive Care Unit (SICU)</td>\n",
       "      <td>2150-05-06 20:44:05</td>\n",
       "      <td>2150-05-15 18:33:48</td>\n",
       "      <td>8.909525</td>\n",
       "      <td>2150-04-30 06:33:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>test</td>\n",
       "      <td>58966412</td>\n",
       "      <td>A Patient was admitted to the ICU in the Surgi...</td>\n",
       "      <td>A male, white, 65 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: As compar...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>The patient received Diazepam (Valium), Famoti...</td>\n",
       "      <td>The patient's diagnoses are Unspecified causes...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6797</th>\n",
       "      <td>13575992</td>\n",
       "      <td>28016813</td>\n",
       "      <td>39983674</td>\n",
       "      <td>13575992_episode1_timeseries.csv</td>\n",
       "      <td>556708b8-f80d5345-d35717bf-6435d670-007dea10</td>\n",
       "      <td>Coronary Care Unit (CCU)</td>\n",
       "      <td>2168-02-12 16:24:20</td>\n",
       "      <td>2168-02-15 15:13:06</td>\n",
       "      <td>2.950532</td>\n",
       "      <td>2168-02-09 11:17:00</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>42217517.0</td>\n",
       "      <td>test</td>\n",
       "      <td>56255444</td>\n",
       "      <td>A Patient was admitted to the ICU in the Coron...</td>\n",
       "      <td>A male, other, 49 years old patient with a wei...</td>\n",
       "      <td>CXR findings: None. CXR impressions: None.</td>\n",
       "      <td>The ECG shows a Sinus rhythm, Left bundle bran...</td>\n",
       "      <td>The patient received Furosemide (Lasix) 250/50...</td>\n",
       "      <td>The patient's diagnoses are Acute on chronic c...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6798 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      subject_id   hadm_id   stay_id                            id_ehr  \\\n",
       "0       18730522  21463350  30004391  18730522_episode2_timeseries.csv   \n",
       "1       12509799  25897223  30004530  12509799_episode1_timeseries.csv   \n",
       "2       10332722  29393377  30005362  10332722_episode1_timeseries.csv   \n",
       "3       18855147  25702658  30005366  18855147_episode2_timeseries.csv   \n",
       "4       15524760  28718343  30006565  15524760_episode1_timeseries.csv   \n",
       "...          ...       ...       ...                               ...   \n",
       "6793    12127320  23231713  39970416  12127320_episode1_timeseries.csv   \n",
       "6794    16438315  21184320  39972274  16438315_episode1_timeseries.csv   \n",
       "6795    14851188  22655854  39972327  14851188_episode1_timeseries.csv   \n",
       "6796    18315784  28656232  39982305  18315784_episode1_timeseries.csv   \n",
       "6797    13575992  28016813  39983674  13575992_episode1_timeseries.csv   \n",
       "\n",
       "                                            id_cxr  \\\n",
       "0     d2bae3a3-3917d71b-f44edcd7-81c7017a-f15288e7   \n",
       "1     b07e6c45-cab5a33c-bb1cade8-a043a627-48572ac5   \n",
       "2     e4e55bf2-3b86fa59-5138badb-62e90dd5-489e3902   \n",
       "3     b361a1e1-b9c3ab9d-c2cc8344-2903cfd8-3888d7b9   \n",
       "4     65dfa87f-d808fe8b-e2c80e3c-f66ae4ff-ea264e9c   \n",
       "...                                            ...   \n",
       "6793  9c8f4dc5-808f29b9-f340457c-52c4ad5e-c2ad0ce4   \n",
       "6794  d91fae99-1f7929e2-d200297e-5569d6c9-d8c345e4   \n",
       "6795  c162887f-d90031ff-a5dae741-f20e7f14-b5022765   \n",
       "6796  4c373404-894fc136-26d5fb93-fe4f8dae-8d5cafbd   \n",
       "6797  556708b8-f80d5345-d35717bf-6435d670-007dea10   \n",
       "\n",
       "                                     last_careunit               intime  \\\n",
       "0               Medical Intensive Care Unit (MICU)  2153-09-05 13:12:00   \n",
       "1     Cardiac Vascular Intensive Care Unit (CVICU)  2165-07-31 09:40:35   \n",
       "2               Medical Intensive Care Unit (MICU)  2156-12-28 21:42:10   \n",
       "3                         Coronary Care Unit (CCU)  2202-12-27 17:36:59   \n",
       "4               Medical Intensive Care Unit (MICU)  2127-07-18 12:44:45   \n",
       "...                                            ...                  ...   \n",
       "6793           Surgical Intensive Care Unit (SICU)  2153-11-12 12:45:31   \n",
       "6794                           Trauma SICU (TSICU)  2185-03-09 01:32:00   \n",
       "6795  Cardiac Vascular Intensive Care Unit (CVICU)  2158-07-26 17:45:10   \n",
       "6796           Surgical Intensive Care Unit (SICU)  2150-05-06 20:44:05   \n",
       "6797                      Coronary Care Unit (CCU)  2168-02-12 16:24:20   \n",
       "\n",
       "                  outtime       los            admittime  ... last_echo_id  \\\n",
       "0     2153-09-13 18:21:18  8.214792  2153-09-05 11:58:00  ...          NaN   \n",
       "1     2165-08-03 16:29:09  3.283727  2165-07-29 06:44:00  ...          NaN   \n",
       "2     2157-01-03 19:16:43  5.898993  2156-12-15 23:03:00  ...          NaN   \n",
       "3     2202-12-31 14:30:31  3.870509  2202-12-27 17:36:00  ...          NaN   \n",
       "4     2127-07-21 20:32:57  3.325139  2127-07-17 22:53:00  ...          NaN   \n",
       "...                   ...       ...                  ...  ...          ...   \n",
       "6793  2153-11-15 20:41:11  3.330324  2153-11-12 06:54:00  ...          NaN   \n",
       "6794  2185-03-17 18:57:06  8.725764  2185-03-09 00:49:00  ...          NaN   \n",
       "6795  2158-07-31 20:10:25  5.100868  2158-07-26 17:44:00  ...          NaN   \n",
       "6796  2150-05-15 18:33:48  8.909525  2150-04-30 06:33:00  ...          NaN   \n",
       "6797  2168-02-15 15:13:06  2.950532  2168-02-09 11:17:00  ...          NaN   \n",
       "\n",
       "     last_ecg_id  split study_id_cxr  \\\n",
       "0            NaN  train     51967233   \n",
       "1     43916976.0  train     55465259   \n",
       "2     40544966.0  train     57947776   \n",
       "3     41783871.0  train     54043642   \n",
       "4            NaN  train     52141331   \n",
       "...          ...    ...          ...   \n",
       "6793  43475719.0   test     58201427   \n",
       "6794         NaN   test     52834029   \n",
       "6795  41452079.0   test     54985656   \n",
       "6796         NaN   test     58966412   \n",
       "6797  42217517.0   test     56255444   \n",
       "\n",
       "                                     demographics_text2  \\\n",
       "0     A Patient was admitted to the ICU in the Medic...   \n",
       "1     A Patient was admitted to the ICU in the Cardi...   \n",
       "2     A Patient was admitted to the ICU in the Medic...   \n",
       "3     A Patient was admitted to the ICU in the Coron...   \n",
       "4     A Patient was admitted to the ICU in the Medic...   \n",
       "...                                                 ...   \n",
       "6793  A Patient was admitted to the ICU in the Surgi...   \n",
       "6794  A Patient was admitted to the ICU in the Traum...   \n",
       "6795  A Patient was admitted to the ICU in the Cardi...   \n",
       "6796  A Patient was admitted to the ICU in the Surgi...   \n",
       "6797  A Patient was admitted to the ICU in the Coron...   \n",
       "\n",
       "                                      demographics_text  \\\n",
       "0     A male, white, 89 years old patient with a wei...   \n",
       "1     A male, white, 63 years old patient with a wei...   \n",
       "2     A male, white, 80 years old patient with a wei...   \n",
       "3     A female, white, 69 years old patient with a w...   \n",
       "4     A male, white, 53 years old patient with a wei...   \n",
       "...                                                 ...   \n",
       "6793  A male, unable to obtain, 72 years old patient...   \n",
       "6794  A female, black/african american, 33 years old...   \n",
       "6795  A male, white, 62 years old patient with a wei...   \n",
       "6796  A male, white, 65 years old patient with a wei...   \n",
       "6797  A male, other, 49 years old patient with a wei...   \n",
       "\n",
       "                                               cxr_text  \\\n",
       "0     CXR findings: None. CXR impressions: Left PICC...   \n",
       "1            CXR findings: None. CXR impressions: None.   \n",
       "2            CXR findings: None. CXR impressions: None.   \n",
       "3            CXR findings: None. CXR impressions: None.   \n",
       "4     CXR findings: None. CXR impressions: Mild card...   \n",
       "...                                                 ...   \n",
       "6793         CXR findings: None. CXR impressions: None.   \n",
       "6794  CXR findings: The endotracheal tube ends 5 cm ...   \n",
       "6795  CXR findings: None. CXR impressions: In compar...   \n",
       "6796  CXR findings: None. CXR impressions: As compar...   \n",
       "6797         CXR findings: None. CXR impressions: None.   \n",
       "\n",
       "                                               ecg_text  \\\n",
       "0                                                   NaN   \n",
       "1     The ECG shows a Sinus rhythm, Left bundle bran...   \n",
       "2     The ECG shows a Sinus tachycardia with PAC(s),...   \n",
       "3     The ECG shows a Sinus rhythm, Possible septal ...   \n",
       "4                                                   NaN   \n",
       "...                                                 ...   \n",
       "6793  The ECG shows a Atrial fibrillation with rapid...   \n",
       "6794                                                NaN   \n",
       "6795  The ECG shows a Sinus rhythm., Possible inferi...   \n",
       "6796                                                NaN   \n",
       "6797  The ECG shows a Sinus rhythm, Left bundle bran...   \n",
       "\n",
       "                                               med_text  \\\n",
       "0     The patient received Dexmedetomidine (Precedex...   \n",
       "1     The patient received Acetaminophen-IV, Calcium...   \n",
       "2     The patient received Heparin Sodium, Heparin S...   \n",
       "3     The patient received Haloperidol (Haldol), Met...   \n",
       "4     The patient received Famotidine (Pepcid), Fent...   \n",
       "...                                                 ...   \n",
       "6793  The patient received Famotidine (Pepcid), Halo...   \n",
       "6794  The patient received Famotidine (Pepcid), Fent...   \n",
       "6795  The patient received Acetaminophen-IV, Calcium...   \n",
       "6796  The patient received Diazepam (Valium), Famoti...   \n",
       "6797  The patient received Furosemide (Lasix) 250/50...   \n",
       "\n",
       "                                               icd_text  \n",
       "0     The patient's diagnoses are Sepsis due to Esch...  \n",
       "1     The patient's diagnoses are Coronary atheroscl...  \n",
       "2     The patient's diagnoses are Urinary tract infe...  \n",
       "3     The patient's diagnoses are Acute myocardial i...  \n",
       "4     The patient's diagnoses are Pneumonitis due to...  \n",
       "...                                                 ...  \n",
       "6793  The patient's diagnoses are Hypertensive encep...  \n",
       "6794  The patient's diagnoses are Cerebral cysts, No...  \n",
       "6795  The patient's diagnoses are Abdominal aneurysm...  \n",
       "6796  The patient's diagnoses are Unspecified causes...  \n",
       "6797  The patient's diagnoses are Acute on chronic c...  \n",
       "\n",
       "[6798 rows x 31 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# train_meta_info = pd.read_csv('data/train_meta_info.csv')\n",
    "# train_meta_info['split'] = 'train'\n",
    "# val_meta_info = pd.read_csv('data/val_meta_info.csv')\n",
    "# val_meta_info['split'] = 'val'\n",
    "# test_meta_info = pd.read_csv('data/test_meta_info.csv')\n",
    "# test_meta_info['split'] = 'test'\n",
    "# all_subjects = pd.concat([train_meta_info, val_meta_info, test_meta_info], ignore_index=True)\n",
    "# all_subjects = all_subjects[[c for c in all_subjects.columns if not c.startswith('Unnamed')]]\n",
    "# all_subjects.to_csv('data/meta.csv')\n",
    "all_subjects = pd.read_csv('data/meta.csv', index_col=0)\n",
    "all_subjects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4ab80011-d412-49fd-8c26-89f7a9f2994c",
   "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>itemid</th>\n",
       "      <th>label</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>linksto</th>\n",
       "      <th>category</th>\n",
       "      <th>unitname</th>\n",
       "      <th>param_type</th>\n",
       "      <th>lownormalvalue</th>\n",
       "      <th>highnormalvalue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>220001</td>\n",
       "      <td>Problem List</td>\n",
       "      <td>Problem List</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>General</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>220003</td>\n",
       "      <td>ICU Admission date</td>\n",
       "      <td>ICU Admission date</td>\n",
       "      <td>datetimeevents</td>\n",
       "      <td>ADT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Date and time</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>220045</td>\n",
       "      <td>Heart Rate</td>\n",
       "      <td>HR</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Routine Vital Signs</td>\n",
       "      <td>bpm</td>\n",
       "      <td>Numeric</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>220046</td>\n",
       "      <td>Heart rate Alarm - High</td>\n",
       "      <td>HR Alarm - High</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Alarms</td>\n",
       "      <td>bpm</td>\n",
       "      <td>Numeric</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>220047</td>\n",
       "      <td>Heart Rate Alarm - Low</td>\n",
       "      <td>HR Alarm - Low</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Alarms</td>\n",
       "      <td>bpm</td>\n",
       "      <td>Numeric</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4009</th>\n",
       "      <td>230077</td>\n",
       "      <td>Documented on DC Summary (PICC)</td>\n",
       "      <td>Documented on DC Summary (PICC)</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>PICC Line Insertion</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4010</th>\n",
       "      <td>230078</td>\n",
       "      <td>Documented on DC Summary (Thora)</td>\n",
       "      <td>Documented on DC Summary (Thora)</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Thoracentesis</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4011</th>\n",
       "      <td>230083</td>\n",
       "      <td>Replacement Fluid (Pre-Filter)</td>\n",
       "      <td>Replacement Fluid (Pre-Filter)</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Dialysis</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4012</th>\n",
       "      <td>230084</td>\n",
       "      <td>Replacement Fluid (Post-Filter)</td>\n",
       "      <td>Replacement Fluid (Post-Filter)</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Dialysis</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4013</th>\n",
       "      <td>230085</td>\n",
       "      <td>Dialysate Fluid</td>\n",
       "      <td>Dialysate Fluid</td>\n",
       "      <td>chartevents</td>\n",
       "      <td>Dialysis</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Text</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4014 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      itemid                             label  \\\n",
       "0     220001                      Problem List   \n",
       "1     220003                ICU Admission date   \n",
       "2     220045                        Heart Rate   \n",
       "3     220046           Heart rate Alarm - High   \n",
       "4     220047            Heart Rate Alarm - Low   \n",
       "...      ...                               ...   \n",
       "4009  230077   Documented on DC Summary (PICC)   \n",
       "4010  230078  Documented on DC Summary (Thora)   \n",
       "4011  230083    Replacement Fluid (Pre-Filter)   \n",
       "4012  230084   Replacement Fluid (Post-Filter)   \n",
       "4013  230085                   Dialysate Fluid   \n",
       "\n",
       "                          abbreviation         linksto             category  \\\n",
       "0                         Problem List     chartevents              General   \n",
       "1                   ICU Admission date  datetimeevents                  ADT   \n",
       "2                                   HR     chartevents  Routine Vital Signs   \n",
       "3                      HR Alarm - High     chartevents               Alarms   \n",
       "4                       HR Alarm - Low     chartevents               Alarms   \n",
       "...                                ...             ...                  ...   \n",
       "4009   Documented on DC Summary (PICC)     chartevents  PICC Line Insertion   \n",
       "4010  Documented on DC Summary (Thora)     chartevents        Thoracentesis   \n",
       "4011    Replacement Fluid (Pre-Filter)     chartevents             Dialysis   \n",
       "4012   Replacement Fluid (Post-Filter)     chartevents             Dialysis   \n",
       "4013                   Dialysate Fluid     chartevents             Dialysis   \n",
       "\n",
       "     unitname     param_type  lownormalvalue  highnormalvalue  \n",
       "0         NaN           Text             NaN              NaN  \n",
       "1         NaN  Date and time             NaN              NaN  \n",
       "2         bpm        Numeric             NaN              NaN  \n",
       "3         bpm        Numeric             NaN              NaN  \n",
       "4         bpm        Numeric             NaN              NaN  \n",
       "...       ...            ...             ...              ...  \n",
       "4009      NaN           Text             NaN              NaN  \n",
       "4010      NaN           Text             NaN              NaN  \n",
       "4011      NaN           Text             NaN              NaN  \n",
       "4012      NaN           Text             NaN              NaN  \n",
       "4013      NaN           Text             NaN              NaN  \n",
       "\n",
       "[4014 rows x 9 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d_item = pd.read_csv('/mnt/sds/sd20i001/malte/data/physionet.org/files/mimiciv/2.2/icu/d_items.csv.gz')\n",
    "d_item"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "301526e0-6494-449a-9c01-20e726d2c53f",
   "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>itemid</th>\n",
       "      <th>label</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>linksto</th>\n",
       "      <th>category</th>\n",
       "      <th>unitname</th>\n",
       "      <th>param_type</th>\n",
       "      <th>lownormalvalue</th>\n",
       "      <th>highnormalvalue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>214</th>\n",
       "      <td>220995</td>\n",
       "      <td>Sodium Bicarbonate 8.4%</td>\n",
       "      <td>Sodium Bicarbonate 8.4%</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mEq</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>278</th>\n",
       "      <td>221261</td>\n",
       "      <td>Abciximab (Reopro)</td>\n",
       "      <td>Abciximab (Reopro)</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>279</th>\n",
       "      <td>221282</td>\n",
       "      <td>Adenosine</td>\n",
       "      <td>Adenosine</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>280</th>\n",
       "      <td>221289</td>\n",
       "      <td>Epinephrine</td>\n",
       "      <td>Epinephrine</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>281</th>\n",
       "      <td>221319</td>\n",
       "      <td>Alteplase (TPA)</td>\n",
       "      <td>Alteplase (TPA)</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3851</th>\n",
       "      <td>229764</td>\n",
       "      <td>Angiotensin II (Giapreza)</td>\n",
       "      <td>Angiotensin II (Giapreza)</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>ng</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3863</th>\n",
       "      <td>229781</td>\n",
       "      <td>Bivalirudin (Angiomax) (Impella)</td>\n",
       "      <td>Bivalirudin (Angiomax) (Impella)</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3919</th>\n",
       "      <td>229861</td>\n",
       "      <td>Sodium Acetate.</td>\n",
       "      <td>Sodium Acetate.</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mEq</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3976</th>\n",
       "      <td>230034</td>\n",
       "      <td>Amiodarone 150/100</td>\n",
       "      <td>Amiodarone 150/100</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>mg</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3985</th>\n",
       "      <td>230044</td>\n",
       "      <td>Heparin Sodium (CRRT-Prefilter)</td>\n",
       "      <td>Heparin Sodium (CRRT-Prefilter)</td>\n",
       "      <td>inputevents</td>\n",
       "      <td>Medications</td>\n",
       "      <td>units</td>\n",
       "      <td>Solution</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>131 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      itemid                             label  \\\n",
       "214   220995           Sodium Bicarbonate 8.4%   \n",
       "278   221261                Abciximab (Reopro)   \n",
       "279   221282                         Adenosine   \n",
       "280   221289                       Epinephrine   \n",
       "281   221319                   Alteplase (TPA)   \n",
       "...      ...                               ...   \n",
       "3851  229764         Angiotensin II (Giapreza)   \n",
       "3863  229781  Bivalirudin (Angiomax) (Impella)   \n",
       "3919  229861                   Sodium Acetate.   \n",
       "3976  230034                Amiodarone 150/100   \n",
       "3985  230044   Heparin Sodium (CRRT-Prefilter)   \n",
       "\n",
       "                          abbreviation      linksto     category unitname  \\\n",
       "214            Sodium Bicarbonate 8.4%  inputevents  Medications      mEq   \n",
       "278                 Abciximab (Reopro)  inputevents  Medications       mg   \n",
       "279                          Adenosine  inputevents  Medications       mg   \n",
       "280                        Epinephrine  inputevents  Medications       mg   \n",
       "281                    Alteplase (TPA)  inputevents  Medications       mg   \n",
       "...                                ...          ...          ...      ...   \n",
       "3851         Angiotensin II (Giapreza)  inputevents  Medications       ng   \n",
       "3863  Bivalirudin (Angiomax) (Impella)  inputevents  Medications       mg   \n",
       "3919                   Sodium Acetate.  inputevents  Medications      mEq   \n",
       "3976                Amiodarone 150/100  inputevents  Medications       mg   \n",
       "3985   Heparin Sodium (CRRT-Prefilter)  inputevents  Medications    units   \n",
       "\n",
       "     param_type  lownormalvalue  highnormalvalue  \n",
       "214    Solution             NaN              NaN  \n",
       "278    Solution             NaN              NaN  \n",
       "279    Solution             NaN              NaN  \n",
       "280    Solution             NaN              NaN  \n",
       "281    Solution             NaN              NaN  \n",
       "...         ...             ...              ...  \n",
       "3851   Solution             NaN              NaN  \n",
       "3863   Solution             NaN              NaN  \n",
       "3919   Solution             NaN              NaN  \n",
       "3976   Solution             NaN              NaN  \n",
       "3985   Solution             NaN              NaN  \n",
       "\n",
       "[131 rows x 9 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "med_d_item = d_item[(d_item['category'] == 'Medications') & (d_item['linksto'] == 'inputevents')]\n",
    "uniqe_med_ditem = med_d_item['itemid'].unique()\n",
    "uniqe_med_ditem\n",
    "med_d_item"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "07dbce2d-3fea-4d64-a6b7-6e46913f2f28",
   "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>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>stay_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>amountuom</th>\n",
       "      <th>...</th>\n",
       "      <th>ordercomponenttypedescription</th>\n",
       "      <th>ordercategorydescription</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>totalamount</th>\n",
       "      <th>totalamountuom</th>\n",
       "      <th>isopenbag</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10000032</td>\n",
       "      <td>29079034</td>\n",
       "      <td>39553978</td>\n",
       "      <td>66056</td>\n",
       "      <td>2180-07-23 21:10:00</td>\n",
       "      <td>2180-07-23 21:11:00</td>\n",
       "      <td>2180-07-23 21:10:00</td>\n",
       "      <td>226452</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Bolus</td>\n",
       "      <td>39.4</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000032</td>\n",
       "      <td>29079034</td>\n",
       "      <td>39553978</td>\n",
       "      <td>88981</td>\n",
       "      <td>2180-07-23 17:00:00</td>\n",
       "      <td>2180-07-23 17:01:00</td>\n",
       "      <td>2180-07-23 18:56:00</td>\n",
       "      <td>226452</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Bolus</td>\n",
       "      <td>39.4</td>\n",
       "      <td>200.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>200.0</td>\n",
       "      <td>200.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10000032</td>\n",
       "      <td>29079034</td>\n",
       "      <td>39553978</td>\n",
       "      <td>88981</td>\n",
       "      <td>2180-07-23 17:00:00</td>\n",
       "      <td>2180-07-23 17:30:00</td>\n",
       "      <td>2180-07-23 17:02:00</td>\n",
       "      <td>220862</td>\n",
       "      <td>49.999999</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Continuous IV</td>\n",
       "      <td>39.4</td>\n",
       "      <td>50.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>50.0</td>\n",
       "      <td>100.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10000032</td>\n",
       "      <td>29079034</td>\n",
       "      <td>39553978</td>\n",
       "      <td>88981</td>\n",
       "      <td>2180-07-23 17:33:00</td>\n",
       "      <td>2180-07-23 18:03:00</td>\n",
       "      <td>2180-07-23 18:16:00</td>\n",
       "      <td>220862</td>\n",
       "      <td>49.999999</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Continuous IV</td>\n",
       "      <td>39.4</td>\n",
       "      <td>50.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>50.0</td>\n",
       "      <td>100.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10000032</td>\n",
       "      <td>29079034</td>\n",
       "      <td>39553978</td>\n",
       "      <td>88981</td>\n",
       "      <td>2180-07-23 18:56:00</td>\n",
       "      <td>2180-07-23 18:57:00</td>\n",
       "      <td>2180-07-23 18:56:00</td>\n",
       "      <td>226452</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Bolus</td>\n",
       "      <td>39.4</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8978888</th>\n",
       "      <td>19999987</td>\n",
       "      <td>23865745</td>\n",
       "      <td>36195440</td>\n",
       "      <td>90295</td>\n",
       "      <td>2145-11-02 23:28:00</td>\n",
       "      <td>2145-11-03 09:32:00</td>\n",
       "      <td>2145-11-02 23:28:00</td>\n",
       "      <td>225942</td>\n",
       "      <td>0.755000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>60.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Paused</td>\n",
       "      <td>2.5</td>\n",
       "      <td>75.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8978889</th>\n",
       "      <td>19999987</td>\n",
       "      <td>23865745</td>\n",
       "      <td>36195440</td>\n",
       "      <td>90295</td>\n",
       "      <td>2145-11-02 23:28:00</td>\n",
       "      <td>2145-11-03 09:32:00</td>\n",
       "      <td>2145-11-02 23:28:00</td>\n",
       "      <td>225943</td>\n",
       "      <td>15.100000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Mixed solution</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>60.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Paused</td>\n",
       "      <td>50.0</td>\n",
       "      <td>1.50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8978890</th>\n",
       "      <td>19999987</td>\n",
       "      <td>23865745</td>\n",
       "      <td>36195440</td>\n",
       "      <td>90295</td>\n",
       "      <td>2145-11-02 23:38:00</td>\n",
       "      <td>2145-11-03 12:55:00</td>\n",
       "      <td>2145-11-03 14:29:00</td>\n",
       "      <td>225158</td>\n",
       "      <td>1000.000037</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Continuous IV</td>\n",
       "      <td>60.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>75.28231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8978891</th>\n",
       "      <td>19999987</td>\n",
       "      <td>23865745</td>\n",
       "      <td>36195440</td>\n",
       "      <td>91879</td>\n",
       "      <td>2145-11-04 20:35:00</td>\n",
       "      <td>2145-11-04 20:36:00</td>\n",
       "      <td>2145-11-04 20:35:00</td>\n",
       "      <td>226452</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Bolus</td>\n",
       "      <td>94.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>200.0</td>\n",
       "      <td>200.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8978892</th>\n",
       "      <td>19999987</td>\n",
       "      <td>23865745</td>\n",
       "      <td>36195440</td>\n",
       "      <td>91879</td>\n",
       "      <td>2145-11-04 20:47:00</td>\n",
       "      <td>2145-11-04 20:54:00</td>\n",
       "      <td>2145-11-04 20:48:00</td>\n",
       "      <td>226089</td>\n",
       "      <td>1.200000</td>\n",
       "      <td>ml</td>\n",
       "      <td>...</td>\n",
       "      <td>Main order parameter</td>\n",
       "      <td>Continuous IV</td>\n",
       "      <td>94.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.2</td>\n",
       "      <td>10.00000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8978893 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         subject_id   hadm_id   stay_id  caregiver_id           starttime  \\\n",
       "0          10000032  29079034  39553978         66056 2180-07-23 21:10:00   \n",
       "1          10000032  29079034  39553978         88981 2180-07-23 17:00:00   \n",
       "2          10000032  29079034  39553978         88981 2180-07-23 17:00:00   \n",
       "3          10000032  29079034  39553978         88981 2180-07-23 17:33:00   \n",
       "4          10000032  29079034  39553978         88981 2180-07-23 18:56:00   \n",
       "...             ...       ...       ...           ...                 ...   \n",
       "8978888    19999987  23865745  36195440         90295 2145-11-02 23:28:00   \n",
       "8978889    19999987  23865745  36195440         90295 2145-11-02 23:28:00   \n",
       "8978890    19999987  23865745  36195440         90295 2145-11-02 23:38:00   \n",
       "8978891    19999987  23865745  36195440         91879 2145-11-04 20:35:00   \n",
       "8978892    19999987  23865745  36195440         91879 2145-11-04 20:47:00   \n",
       "\n",
       "                     endtime            storetime  itemid       amount  \\\n",
       "0        2180-07-23 21:11:00  2180-07-23 21:10:00  226452   100.000000   \n",
       "1        2180-07-23 17:01:00  2180-07-23 18:56:00  226452   200.000000   \n",
       "2        2180-07-23 17:30:00  2180-07-23 17:02:00  220862    49.999999   \n",
       "3        2180-07-23 18:03:00  2180-07-23 18:16:00  220862    49.999999   \n",
       "4        2180-07-23 18:57:00  2180-07-23 18:56:00  226452   100.000000   \n",
       "...                      ...                  ...     ...          ...   \n",
       "8978888  2145-11-03 09:32:00  2145-11-02 23:28:00  225942     0.755000   \n",
       "8978889  2145-11-03 09:32:00  2145-11-02 23:28:00  225943    15.100000   \n",
       "8978890  2145-11-03 12:55:00  2145-11-03 14:29:00  225158  1000.000037   \n",
       "8978891  2145-11-04 20:36:00  2145-11-04 20:35:00  226452   200.000000   \n",
       "8978892  2145-11-04 20:54:00  2145-11-04 20:48:00  226089     1.200000   \n",
       "\n",
       "        amountuom  ...  ordercomponenttypedescription  \\\n",
       "0              ml  ...           Main order parameter   \n",
       "1              ml  ...           Main order parameter   \n",
       "2              ml  ...           Main order parameter   \n",
       "3              ml  ...           Main order parameter   \n",
       "4              ml  ...           Main order parameter   \n",
       "...           ...  ...                            ...   \n",
       "8978888        mg  ...           Main order parameter   \n",
       "8978889        ml  ...                 Mixed solution   \n",
       "8978890        ml  ...           Main order parameter   \n",
       "8978891        ml  ...           Main order parameter   \n",
       "8978892        ml  ...           Main order parameter   \n",
       "\n",
       "        ordercategorydescription  patientweight  totalamount totalamountuom  \\\n",
       "0                          Bolus           39.4        100.0             ml   \n",
       "1                          Bolus           39.4        200.0             ml   \n",
       "2                  Continuous IV           39.4         50.0             ml   \n",
       "3                  Continuous IV           39.4         50.0             ml   \n",
       "4                          Bolus           39.4        100.0             ml   \n",
       "...                          ...            ...          ...            ...   \n",
       "8978888           Continuous Med           60.0         50.0             ml   \n",
       "8978889           Continuous Med           60.0         50.0             ml   \n",
       "8978890            Continuous IV           60.0       1000.0             ml   \n",
       "8978891                    Bolus           94.0        200.0             ml   \n",
       "8978892            Continuous IV           94.0          1.2             ml   \n",
       "\n",
       "        isopenbag continueinnextdept statusdescription  originalamount  \\\n",
       "0               0                  0   FinishedRunning           100.0   \n",
       "1               0                  0   FinishedRunning           200.0   \n",
       "2               0                  0   FinishedRunning            50.0   \n",
       "3               0                  0   FinishedRunning            50.0   \n",
       "4               0                  0   FinishedRunning           100.0   \n",
       "...           ...                ...               ...             ...   \n",
       "8978888         0                  0            Paused             2.5   \n",
       "8978889         0                  0            Paused            50.0   \n",
       "8978890         0                  0   FinishedRunning          1000.0   \n",
       "8978891         0                  0   FinishedRunning           200.0   \n",
       "8978892         0                  0   FinishedRunning             1.2   \n",
       "\n",
       "         originalrate  \n",
       "0           100.00000  \n",
       "1           200.00000  \n",
       "2           100.00000  \n",
       "3           100.00000  \n",
       "4           100.00000  \n",
       "...               ...  \n",
       "8978888      75.00000  \n",
       "8978889       1.50000  \n",
       "8978890      75.28231  \n",
       "8978891     200.00000  \n",
       "8978892      10.00000  \n",
       "\n",
       "[8978893 rows x 26 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputevents = pd.read_csv('/mnt/sds/sd20i001/malte/data/physionet.org/files/mimiciv/2.2/icu/inputevents.csv.gz')\n",
    "inputevents['starttime'] = pd.to_datetime(inputevents['starttime'])\n",
    "\n",
    "inputevents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a21f8334-8aef-4ded-bf0b-c598d664ef76",
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered_item_ids = med_d_item['itemid'].unique()\n",
    "filtered_stay_ids = all_subjects['stay_id'].unique()\n",
    "filtered_inputevents = inputevents[(inputevents['itemid'].isin(filtered_item_ids)) & (inputevents['stay_id'].isin(filtered_stay_ids))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "e7df86b3-0f3b-4410-8407-6d3e79dfff9c",
   "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>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>stay_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>amountuom</th>\n",
       "      <th>...</th>\n",
       "      <th>ordercategorydescription</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>totalamount</th>\n",
       "      <th>totalamountuom</th>\n",
       "      <th>isopenbag</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>37510196</td>\n",
       "      <td>10221</td>\n",
       "      <td>2131-01-18 21:04:00</td>\n",
       "      <td>2131-01-19 02:50:00</td>\n",
       "      <td>2131-01-19 08:52:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>798.499966</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>65.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>798.500000</td>\n",
       "      <td>35.496826</td>\n",
       "      <td>Propofol</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>37510196</td>\n",
       "      <td>10221</td>\n",
       "      <td>2131-01-19 21:06:00</td>\n",
       "      <td>2131-01-19 21:07:00</td>\n",
       "      <td>2131-01-19 21:06:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>65.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>37510196</td>\n",
       "      <td>10221</td>\n",
       "      <td>2131-01-19 21:06:00</td>\n",
       "      <td>2131-01-19 22:30:00</td>\n",
       "      <td>2131-01-19 21:06:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>65.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>98.900002</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>37510196</td>\n",
       "      <td>18012</td>\n",
       "      <td>2131-01-16 07:57:00</td>\n",
       "      <td>2131-01-16 15:49:00</td>\n",
       "      <td>2131-01-16 09:11:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>612.443620</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>65.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>1000.000061</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>Propofol</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>37510196</td>\n",
       "      <td>18012</td>\n",
       "      <td>2131-01-16 09:09:00</td>\n",
       "      <td>2131-01-16 14:01:00</td>\n",
       "      <td>2131-01-16 09:08:00</td>\n",
       "      <td>225148</td>\n",
       "      <td>72.120461</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Continuous Med</td>\n",
       "      <td>65.0</td>\n",
       "      <td>500.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>213.644531</td>\n",
       "      <td>0.229000</td>\n",
       "      <td>Bivalirudin (Angiomax)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672377</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>32336619</td>\n",
       "      <td>87525</td>\n",
       "      <td>2148-11-25 07:20:00</td>\n",
       "      <td>2148-11-25 07:21:00</td>\n",
       "      <td>2148-11-25 09:20:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672378</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>32336619</td>\n",
       "      <td>87525</td>\n",
       "      <td>2148-11-25 08:09:00</td>\n",
       "      <td>2148-11-25 08:10:00</td>\n",
       "      <td>2148-11-25 08:09:00</td>\n",
       "      <td>225975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>dose</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672379</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>32336619</td>\n",
       "      <td>87525</td>\n",
       "      <td>2148-11-25 09:20:00</td>\n",
       "      <td>2148-11-25 09:21:00</td>\n",
       "      <td>2148-11-25 09:20:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672380</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>32336619</td>\n",
       "      <td>87525</td>\n",
       "      <td>2148-11-25 13:56:00</td>\n",
       "      <td>2148-11-25 13:57:00</td>\n",
       "      <td>2148-11-25 13:56:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672381</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>32336619</td>\n",
       "      <td>87525</td>\n",
       "      <td>2148-11-25 15:47:00</td>\n",
       "      <td>2148-11-25 15:48:00</td>\n",
       "      <td>2148-11-25 15:47:00</td>\n",
       "      <td>225975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>dose</td>\n",
       "      <td>...</td>\n",
       "      <td>Drug Push</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>672382 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        subject_id   hadm_id   stay_id  caregiver_id           starttime  \\\n",
       "0         10001884  26184834  37510196         10221 2131-01-18 21:04:00   \n",
       "1         10001884  26184834  37510196         10221 2131-01-19 21:06:00   \n",
       "2         10001884  26184834  37510196         10221 2131-01-19 21:06:00   \n",
       "3         10001884  26184834  37510196         18012 2131-01-16 07:57:00   \n",
       "4         10001884  26184834  37510196         18012 2131-01-16 09:09:00   \n",
       "...            ...       ...       ...           ...                 ...   \n",
       "672377    19999442  26785317  32336619         87525 2148-11-25 07:20:00   \n",
       "672378    19999442  26785317  32336619         87525 2148-11-25 08:09:00   \n",
       "672379    19999442  26785317  32336619         87525 2148-11-25 09:20:00   \n",
       "672380    19999442  26785317  32336619         87525 2148-11-25 13:56:00   \n",
       "672381    19999442  26785317  32336619         87525 2148-11-25 15:47:00   \n",
       "\n",
       "                    endtime            storetime  itemid      amount  \\\n",
       "0       2131-01-19 02:50:00  2131-01-19 08:52:00  222168  798.499966   \n",
       "1       2131-01-19 21:07:00  2131-01-19 21:06:00  225154    1.000000   \n",
       "2       2131-01-19 22:30:00  2131-01-19 21:06:00  225154    2.800000   \n",
       "3       2131-01-16 15:49:00  2131-01-16 09:11:00  222168  612.443620   \n",
       "4       2131-01-16 14:01:00  2131-01-16 09:08:00  225148   72.120461   \n",
       "...                     ...                  ...     ...         ...   \n",
       "672377  2148-11-25 07:21:00  2148-11-25 09:20:00  221833    0.500000   \n",
       "672378  2148-11-25 08:10:00  2148-11-25 08:09:00  225975    1.000000   \n",
       "672379  2148-11-25 09:21:00  2148-11-25 09:20:00  221833    0.500000   \n",
       "672380  2148-11-25 13:57:00  2148-11-25 13:56:00  221833    0.500000   \n",
       "672381  2148-11-25 15:48:00  2148-11-25 15:47:00  225975    1.000000   \n",
       "\n",
       "       amountuom  ...  ordercategorydescription patientweight  totalamount  \\\n",
       "0             mg  ...            Continuous Med          65.0        100.0   \n",
       "1             mg  ...                 Drug Push          65.0          NaN   \n",
       "2             mg  ...            Continuous Med          65.0        100.0   \n",
       "3             mg  ...            Continuous Med          65.0        100.0   \n",
       "4             mg  ...            Continuous Med          65.0        500.0   \n",
       "...          ...  ...                       ...           ...          ...   \n",
       "672377        mg  ...                 Drug Push         107.5          NaN   \n",
       "672378      dose  ...                 Drug Push         107.5          NaN   \n",
       "672379        mg  ...                 Drug Push         107.5          NaN   \n",
       "672380        mg  ...                 Drug Push         107.5          NaN   \n",
       "672381      dose  ...                 Drug Push         107.5          NaN   \n",
       "\n",
       "        totalamountuom isopenbag continueinnextdept statusdescription  \\\n",
       "0                   ml         0                  0   FinishedRunning   \n",
       "1                  NaN         0                  0   FinishedRunning   \n",
       "2                   ml         0                  0   ChangeDose/Rate   \n",
       "3                   ml         0                  0   ChangeDose/Rate   \n",
       "4                   ml         0                  0   ChangeDose/Rate   \n",
       "...                ...       ...                ...               ...   \n",
       "672377             NaN         0                  0   FinishedRunning   \n",
       "672378             NaN         0                  0   FinishedRunning   \n",
       "672379             NaN         0                  0   FinishedRunning   \n",
       "672380             NaN         0                  0   FinishedRunning   \n",
       "672381             NaN         0                  0   FinishedRunning   \n",
       "\n",
       "       originalamount  originalrate                         label  \n",
       "0          798.500000     35.496826                      Propofol  \n",
       "1            1.000000      1.000000              Morphine Sulfate  \n",
       "2           98.900002      2.000000              Morphine Sulfate  \n",
       "3         1000.000061     20.000000                      Propofol  \n",
       "4          213.644531      0.229000        Bivalirudin (Angiomax)  \n",
       "...               ...           ...                           ...  \n",
       "672377       0.500000      0.500000      Hydromorphone (Dilaudid)  \n",
       "672378       1.000000      1.000000  Heparin Sodium (Prophylaxis)  \n",
       "672379       0.500000      0.500000      Hydromorphone (Dilaudid)  \n",
       "672380       0.500000      0.500000      Hydromorphone (Dilaudid)  \n",
       "672381       1.000000      1.000000  Heparin Sodium (Prophylaxis)  \n",
       "\n",
       "[672382 rows x 27 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_inputevents = filtered_inputevents.merge(d_item[['itemid', 'label']], on='itemid', how='left')\n",
    "merged_inputevents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "e92d9d3c-4b7d-40b9-a02a-430fa7d93aad",
   "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>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>amountuom</th>\n",
       "      <th>rate</th>\n",
       "      <th>...</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>totalamount</th>\n",
       "      <th>totalamountuom</th>\n",
       "      <th>isopenbag</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "      <th>label</th>\n",
       "      <th>time_since_intime</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stay_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>37510196</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>29855</td>\n",
       "      <td>2131-01-12 23:43:00</td>\n",
       "      <td>2131-01-12 23:44:00</td>\n",
       "      <td>2131-01-12 23:43:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>43.381944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37510196</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 08:48:00</td>\n",
       "      <td>2131-01-12 08:49:00</td>\n",
       "      <td>2131-01-12 08:48:00</td>\n",
       "      <td>225910</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>dose</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Pantoprazole (Protonix)</td>\n",
       "      <td>28.465278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37510196</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 11:17:00</td>\n",
       "      <td>2131-01-12 17:23:00</td>\n",
       "      <td>2131-01-12 11:28:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>475.092154</td>\n",
       "      <td>mg</td>\n",
       "      <td>19.970248</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Stopped</td>\n",
       "      <td>663.311768</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>30.948611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37510196</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 11:22:00</td>\n",
       "      <td>2131-01-12 12:42:00</td>\n",
       "      <td>2131-01-12 12:29:00</td>\n",
       "      <td>225152</td>\n",
       "      <td>1394.543152</td>\n",
       "      <td>units</td>\n",
       "      <td>1045.907349</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1394.542969</td>\n",
       "      <td>1040.969727</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>31.031944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37510196</th>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 12:42:00</td>\n",
       "      <td>2131-01-12 18:11:00</td>\n",
       "      <td>2131-01-12 12:29:00</td>\n",
       "      <td>225152</td>\n",
       "      <td>5509.341850</td>\n",
       "      <td>units</td>\n",
       "      <td>1004.743164</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>25000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>32.365278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32336619</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 00:16:00</td>\n",
       "      <td>2148-11-20 01:13:00</td>\n",
       "      <td>2148-11-20 01:42:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>354.746839</td>\n",
       "      <td>mg</td>\n",
       "      <td>57.894222</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>60.000008</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>9.871389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32336619</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 01:13:00</td>\n",
       "      <td>2148-11-20 02:56:00</td>\n",
       "      <td>2148-11-20 01:42:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>427.782282</td>\n",
       "      <td>mg</td>\n",
       "      <td>38.634662</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>643.750000</td>\n",
       "      <td>40.000004</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>10.821389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32336619</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 02:56:00</td>\n",
       "      <td>2148-11-20 04:15:00</td>\n",
       "      <td>2148-11-20 08:38:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>206.470988</td>\n",
       "      <td>mg</td>\n",
       "      <td>24.312157</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>206.470993</td>\n",
       "      <td>25.110836</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>12.538056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32336619</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-21 00:00:00</td>\n",
       "      <td>2148-11-21 00:01:00</td>\n",
       "      <td>2148-11-21 04:10:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "      <td>33.604722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32336619</th>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-21 03:30:00</td>\n",
       "      <td>2148-11-21 03:31:00</td>\n",
       "      <td>2148-11-21 04:10:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "      <td>37.104722</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>268034 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          subject_id   hadm_id  caregiver_id           starttime  \\\n",
       "stay_id                                                            \n",
       "37510196    10001884  26184834         29855 2131-01-12 23:43:00   \n",
       "37510196    10001884  26184834         36518 2131-01-12 08:48:00   \n",
       "37510196    10001884  26184834         36518 2131-01-12 11:17:00   \n",
       "37510196    10001884  26184834         36518 2131-01-12 11:22:00   \n",
       "37510196    10001884  26184834         36518 2131-01-12 12:42:00   \n",
       "...              ...       ...           ...                 ...   \n",
       "32336619    19999442  26785317         59028 2148-11-20 00:16:00   \n",
       "32336619    19999442  26785317         59028 2148-11-20 01:13:00   \n",
       "32336619    19999442  26785317         59028 2148-11-20 02:56:00   \n",
       "32336619    19999442  26785317         59028 2148-11-21 00:00:00   \n",
       "32336619    19999442  26785317         59028 2148-11-21 03:30:00   \n",
       "\n",
       "                      endtime            storetime  itemid       amount  \\\n",
       "stay_id                                                                   \n",
       "37510196  2131-01-12 23:44:00  2131-01-12 23:43:00  225154     1.000000   \n",
       "37510196  2131-01-12 08:49:00  2131-01-12 08:48:00  225910     1.000000   \n",
       "37510196  2131-01-12 17:23:00  2131-01-12 11:28:00  222168   475.092154   \n",
       "37510196  2131-01-12 12:42:00  2131-01-12 12:29:00  225152  1394.543152   \n",
       "37510196  2131-01-12 18:11:00  2131-01-12 12:29:00  225152  5509.341850   \n",
       "...                       ...                  ...     ...          ...   \n",
       "32336619  2148-11-20 01:13:00  2148-11-20 01:42:00  222168   354.746839   \n",
       "32336619  2148-11-20 02:56:00  2148-11-20 01:42:00  222168   427.782282   \n",
       "32336619  2148-11-20 04:15:00  2148-11-20 08:38:00  222168   206.470988   \n",
       "32336619  2148-11-21 00:01:00  2148-11-21 04:10:00  221833     1.000000   \n",
       "32336619  2148-11-21 03:31:00  2148-11-21 04:10:00  221833     1.000000   \n",
       "\n",
       "         amountuom         rate  ... patientweight  totalamount  \\\n",
       "stay_id                          ...                              \n",
       "37510196        mg          NaN  ...          65.0          NaN   \n",
       "37510196      dose          NaN  ...          65.0         10.0   \n",
       "37510196        mg    19.970248  ...          65.0        100.0   \n",
       "37510196     units  1045.907349  ...          65.0        250.0   \n",
       "37510196     units  1004.743164  ...          65.0        250.0   \n",
       "...            ...          ...  ...           ...          ...   \n",
       "32336619        mg    57.894222  ...         107.5        100.0   \n",
       "32336619        mg    38.634662  ...         107.5        100.0   \n",
       "32336619        mg    24.312157  ...         107.5        100.0   \n",
       "32336619        mg          NaN  ...         107.5          NaN   \n",
       "32336619        mg          NaN  ...         107.5          NaN   \n",
       "\n",
       "          totalamountuom isopenbag continueinnextdept statusdescription  \\\n",
       "stay_id                                                                   \n",
       "37510196             NaN         0                  0   FinishedRunning   \n",
       "37510196              ml         0                  0   FinishedRunning   \n",
       "37510196              ml         0                  0           Stopped   \n",
       "37510196              ml         0                  0   FinishedRunning   \n",
       "37510196              ml         0                  0   ChangeDose/Rate   \n",
       "...                  ...       ...                ...               ...   \n",
       "32336619              ml         0                  0   ChangeDose/Rate   \n",
       "32336619              ml         0                  0   ChangeDose/Rate   \n",
       "32336619              ml         0                  0   FinishedRunning   \n",
       "32336619             NaN         0                  0   FinishedRunning   \n",
       "32336619             NaN         0                  0   FinishedRunning   \n",
       "\n",
       "         originalamount  originalrate                     label  \\\n",
       "stay_id                                                           \n",
       "37510196       1.000000      1.000000          Morphine Sulfate   \n",
       "37510196       1.000000      1.000000   Pantoprazole (Protonix)   \n",
       "37510196     663.311768     20.000000                  Propofol   \n",
       "37510196    1394.542969   1040.969727            Heparin Sodium   \n",
       "37510196   25000.000000   1000.000000            Heparin Sodium   \n",
       "...                 ...           ...                       ...   \n",
       "32336619    1000.000000     60.000008                  Propofol   \n",
       "32336619     643.750000     40.000004                  Propofol   \n",
       "32336619     206.470993     25.110836                  Propofol   \n",
       "32336619       1.000000      1.000000  Hydromorphone (Dilaudid)   \n",
       "32336619       1.000000      1.000000  Hydromorphone (Dilaudid)   \n",
       "\n",
       "         time_since_intime  \n",
       "stay_id                     \n",
       "37510196         43.381944  \n",
       "37510196         28.465278  \n",
       "37510196         30.948611  \n",
       "37510196         31.031944  \n",
       "37510196         32.365278  \n",
       "...                    ...  \n",
       "32336619          9.871389  \n",
       "32336619         10.821389  \n",
       "32336619         12.538056  \n",
       "32336619         33.604722  \n",
       "32336619         37.104722  \n",
       "\n",
       "[268034 rows x 27 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_subjects['intime'] = pd.to_datetime(all_subjects['intime'])\n",
    "merged_inputevents['time_since_intime'] = merged_inputevents.apply(\n",
    "    lambda row: (row['starttime'] - all_subjects.loc[all_subjects['stay_id'] == row['stay_id'], 'intime'].values[0]).total_seconds() / 3600,\n",
    "    axis=1\n",
    ")\n",
    "\n",
    "meds_within_48h = merged_inputevents[merged_inputevents['time_since_intime'] <= 48]\n",
    "meds_within_48h.set_index('stay_id', inplace=True)\n",
    "meds_within_48h\n",
    "# meds_within_48h_unique = meds_within_48h.drop_duplicates(subset=['stay_id', 'label'])\n",
    "# medication_dict = meds_within_48h_unique.groupby('stay_id')['label'].apply(list).to_dict()\n",
    "\n",
    "# patient_weight_dict = meds_within_48h.groupby('stay_id')['patientweight'].last().to_dict()\n",
    "\n",
    "# all_subjects['medications'] = all_subjects['stay_id'].map(medication_dict)\n",
    "\n",
    "# all_subjects['patientweight'] = all_subjects['stay_id'].map(patient_weight_dict)\n",
    "\n",
    "# all_subjects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7a82d525",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# med_categories = {\n",
    "#     \"Antiarrhythmics\": [\n",
    "#         \"Amiodarone 600/500\",\n",
    "#         \"Amiodarone\",\n",
    "#         \"Lidocaine\",\n",
    "#         \"Adenosine\",\n",
    "#         \"Procainamide\"\n",
    "#     ],\n",
    "#     \"Electrolyte Supplements\": [\n",
    "#         \"Potassium Chloride\",\n",
    "#         # \"KCL (Bolus)\",\n",
    "#         # \"Magnesium Sulfate\",\n",
    "#         \"Calcium Gluconate\",\n",
    "#         # \"Magnesium Sulfate (Bolus)\",\n",
    "#         \"Calcium Gluconate (CRRT)\",\n",
    "#         # \"KCl (CRRT)\",\n",
    "#         \"K Phos\",\n",
    "#         \"Na Phos\",\n",
    "#         # \"Magnesium Sulfate (OB-GYN)\"\n",
    "#     ],\n",
    "#     \"Beta-Blockers\": [\n",
    "#         \"Metoprolol\",\n",
    "#         \"Labetalol\",\n",
    "#         \"Esmolol\"\n",
    "#     ],\n",
    "#     \"Diuretics\": [\n",
    "#         \"Furosemide (Lasix)\",\n",
    "#         \"Furosemide (Lasix) 250/50\",\n",
    "#         \"Mannitol\"\n",
    "#     ],\n",
    "#     \"Cardiac Glycosides\": [\n",
    "#         \"Digoxin (Lanoxin)\"\n",
    "#     ],\n",
    "#     \"Insulins\": [\n",
    "#         \"Insulin - Regular\",\n",
    "#         \"Insulin - Humalog\",\n",
    "#         \"Insulin - Glargine\",\n",
    "#         \"Insulin - NPH\",\n",
    "#         \"Insulin - 70/30\",\n",
    "#         \"Insulin - Novolog\",\n",
    "#         \"Insulin - Humalog 75/25\"\n",
    "#     ],\n",
    "#     \"Opioid Analgesics\": [\n",
    "#         \"Fentanyl\",\n",
    "#         \"Fentanyl (Concentrate)\",\n",
    "#         \"Hydromorphone (Dilaudid)\",\n",
    "#         \"Morphine Sulfate\",\n",
    "#         \"Meperidine (Demerol)\",\n",
    "#         \"Methadone Hydrochloride\"\n",
    "#     ],\n",
    "#     \"Sedatives and Anesthetics\": [\n",
    "#         \"Propofol\",\n",
    "#         \"Midazolam (Versed)\",\n",
    "#         \"Dexmedetomidine (Precedex)\",\n",
    "#         \"Lorazepam (Ativan)\",\n",
    "#         \"Ketamine\",\n",
    "#         \"Diazepam (Valium)\",\n",
    "#         \"Pentobarbital\"\n",
    "#     ],\n",
    "#     \"Thrombolytics\": [\n",
    "#         \"Alteplase (TPA)\"\n",
    "#     ],\n",
    "#     \"Vasopressors and Inotropes\": [\n",
    "#         \"Norepinephrine\",\n",
    "#         \"Phenylephrine\",\n",
    "#         \"Epinephrine\",\n",
    "#         \"Dopamine\",\n",
    "#         \"Vasopressin\",\n",
    "#         \"Milrinone\",\n",
    "#         \"Dobutamine\",\n",
    "#         \"Isuprel\"\n",
    "#     ],\n",
    "# }\n",
    "\n",
    "# import json\n",
    "# with open('med_categories.json','r') as f:\n",
    "#     med_categories = json.load(f)\n",
    "med_categories_t = {v: k for k, values in med_categories.items() for v in values}\n",
    "len(med_categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6aa54aff",
   "metadata": {},
   "outputs": [],
   "source": [
    "# import json\n",
    "# with open('meds.json', 'w') as f:\n",
    "#     json.dump(med_category_assignment_t,f, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bf06f1d8",
   "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>stay_id</th>\n",
       "      <th>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>amountuom</th>\n",
       "      <th>...</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>totalamount</th>\n",
       "      <th>totalamountuom</th>\n",
       "      <th>isopenbag</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "      <th>label</th>\n",
       "      <th>time_since_intime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>37510196</td>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>29855</td>\n",
       "      <td>2131-01-12 23:43:00</td>\n",
       "      <td>2131-01-12 23:44:00</td>\n",
       "      <td>2131-01-12 23:43:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>43.381944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37510196</td>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 08:48:00</td>\n",
       "      <td>2131-01-12 08:49:00</td>\n",
       "      <td>2131-01-12 08:48:00</td>\n",
       "      <td>225910</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>dose</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Pantoprazole (Protonix)</td>\n",
       "      <td>28.465278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>37510196</td>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 11:17:00</td>\n",
       "      <td>2131-01-12 17:23:00</td>\n",
       "      <td>2131-01-12 11:28:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>475.092154</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Stopped</td>\n",
       "      <td>663.311768</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>30.948611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>37510196</td>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 11:22:00</td>\n",
       "      <td>2131-01-12 12:42:00</td>\n",
       "      <td>2131-01-12 12:29:00</td>\n",
       "      <td>225152</td>\n",
       "      <td>1394.543152</td>\n",
       "      <td>units</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1394.542969</td>\n",
       "      <td>1040.969727</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>31.031944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>37510196</td>\n",
       "      <td>10001884</td>\n",
       "      <td>26184834</td>\n",
       "      <td>36518</td>\n",
       "      <td>2131-01-12 12:42:00</td>\n",
       "      <td>2131-01-12 18:11:00</td>\n",
       "      <td>2131-01-12 12:29:00</td>\n",
       "      <td>225152</td>\n",
       "      <td>5509.341850</td>\n",
       "      <td>units</td>\n",
       "      <td>...</td>\n",
       "      <td>65.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>25000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>32.365278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268029</th>\n",
       "      <td>32336619</td>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 00:16:00</td>\n",
       "      <td>2148-11-20 01:13:00</td>\n",
       "      <td>2148-11-20 01:42:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>354.746839</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>60.000008</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>9.871389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268030</th>\n",
       "      <td>32336619</td>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 01:13:00</td>\n",
       "      <td>2148-11-20 02:56:00</td>\n",
       "      <td>2148-11-20 01:42:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>427.782282</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>643.750000</td>\n",
       "      <td>40.000004</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>10.821389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268031</th>\n",
       "      <td>32336619</td>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-20 02:56:00</td>\n",
       "      <td>2148-11-20 04:15:00</td>\n",
       "      <td>2148-11-20 08:38:00</td>\n",
       "      <td>222168</td>\n",
       "      <td>206.470988</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>100.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>206.470993</td>\n",
       "      <td>25.110836</td>\n",
       "      <td>Propofol</td>\n",
       "      <td>12.538056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268032</th>\n",
       "      <td>32336619</td>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-21 00:00:00</td>\n",
       "      <td>2148-11-21 00:01:00</td>\n",
       "      <td>2148-11-21 04:10:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "      <td>33.604722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268033</th>\n",
       "      <td>32336619</td>\n",
       "      <td>19999442</td>\n",
       "      <td>26785317</td>\n",
       "      <td>59028</td>\n",
       "      <td>2148-11-21 03:30:00</td>\n",
       "      <td>2148-11-21 03:31:00</td>\n",
       "      <td>2148-11-21 04:10:00</td>\n",
       "      <td>221833</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>107.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>Hydromorphone (Dilaudid)</td>\n",
       "      <td>37.104722</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>268034 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         stay_id  subject_id   hadm_id  caregiver_id            starttime  \\\n",
       "0       37510196    10001884  26184834         29855  2131-01-12 23:43:00   \n",
       "1       37510196    10001884  26184834         36518  2131-01-12 08:48:00   \n",
       "2       37510196    10001884  26184834         36518  2131-01-12 11:17:00   \n",
       "3       37510196    10001884  26184834         36518  2131-01-12 11:22:00   \n",
       "4       37510196    10001884  26184834         36518  2131-01-12 12:42:00   \n",
       "...          ...         ...       ...           ...                  ...   \n",
       "268029  32336619    19999442  26785317         59028  2148-11-20 00:16:00   \n",
       "268030  32336619    19999442  26785317         59028  2148-11-20 01:13:00   \n",
       "268031  32336619    19999442  26785317         59028  2148-11-20 02:56:00   \n",
       "268032  32336619    19999442  26785317         59028  2148-11-21 00:00:00   \n",
       "268033  32336619    19999442  26785317         59028  2148-11-21 03:30:00   \n",
       "\n",
       "                    endtime            storetime  itemid       amount  \\\n",
       "0       2131-01-12 23:44:00  2131-01-12 23:43:00  225154     1.000000   \n",
       "1       2131-01-12 08:49:00  2131-01-12 08:48:00  225910     1.000000   \n",
       "2       2131-01-12 17:23:00  2131-01-12 11:28:00  222168   475.092154   \n",
       "3       2131-01-12 12:42:00  2131-01-12 12:29:00  225152  1394.543152   \n",
       "4       2131-01-12 18:11:00  2131-01-12 12:29:00  225152  5509.341850   \n",
       "...                     ...                  ...     ...          ...   \n",
       "268029  2148-11-20 01:13:00  2148-11-20 01:42:00  222168   354.746839   \n",
       "268030  2148-11-20 02:56:00  2148-11-20 01:42:00  222168   427.782282   \n",
       "268031  2148-11-20 04:15:00  2148-11-20 08:38:00  222168   206.470988   \n",
       "268032  2148-11-21 00:01:00  2148-11-21 04:10:00  221833     1.000000   \n",
       "268033  2148-11-21 03:31:00  2148-11-21 04:10:00  221833     1.000000   \n",
       "\n",
       "       amountuom  ...  patientweight totalamount  totalamountuom  isopenbag  \\\n",
       "0             mg  ...           65.0         NaN             NaN          0   \n",
       "1           dose  ...           65.0        10.0              ml          0   \n",
       "2             mg  ...           65.0       100.0              ml          0   \n",
       "3          units  ...           65.0       250.0              ml          0   \n",
       "4          units  ...           65.0       250.0              ml          0   \n",
       "...          ...  ...            ...         ...             ...        ...   \n",
       "268029        mg  ...          107.5       100.0              ml          0   \n",
       "268030        mg  ...          107.5       100.0              ml          0   \n",
       "268031        mg  ...          107.5       100.0              ml          0   \n",
       "268032        mg  ...          107.5         NaN             NaN          0   \n",
       "268033        mg  ...          107.5         NaN             NaN          0   \n",
       "\n",
       "       continueinnextdept statusdescription originalamount originalrate  \\\n",
       "0                       0   FinishedRunning       1.000000     1.000000   \n",
       "1                       0   FinishedRunning       1.000000     1.000000   \n",
       "2                       0           Stopped     663.311768    20.000000   \n",
       "3                       0   FinishedRunning    1394.542969  1040.969727   \n",
       "4                       0   ChangeDose/Rate   25000.000000  1000.000000   \n",
       "...                   ...               ...            ...          ...   \n",
       "268029                  0   ChangeDose/Rate    1000.000000    60.000008   \n",
       "268030                  0   ChangeDose/Rate     643.750000    40.000004   \n",
       "268031                  0   FinishedRunning     206.470993    25.110836   \n",
       "268032                  0   FinishedRunning       1.000000     1.000000   \n",
       "268033                  0   FinishedRunning       1.000000     1.000000   \n",
       "\n",
       "                           label  time_since_intime  \n",
       "0               Morphine Sulfate          43.381944  \n",
       "1        Pantoprazole (Protonix)          28.465278  \n",
       "2                       Propofol          30.948611  \n",
       "3                 Heparin Sodium          31.031944  \n",
       "4                 Heparin Sodium          32.365278  \n",
       "...                          ...                ...  \n",
       "268029                  Propofol           9.871389  \n",
       "268030                  Propofol          10.821389  \n",
       "268031                  Propofol          12.538056  \n",
       "268032  Hydromorphone (Dilaudid)          33.604722  \n",
       "268033  Hydromorphone (Dilaudid)          37.104722  \n",
       "\n",
       "[268034 rows x 28 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# meds_within_48h.to_csv('data/meds_within_48h.csv')\n",
    "meds_within_48h = pd.read_csv('/mnt/hdd/data/MMMedViT_data/data//meds_within_48h.csv')\n",
    "meds_within_48h"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "46a898b8",
   "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>index</th>\n",
       "      <th>stay_id</th>\n",
       "      <th>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>...</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "      <th>label</th>\n",
       "      <th>time_since_intime</th>\n",
       "      <th>cumulative_amount</th>\n",
       "      <th>percentile_95</th>\n",
       "      <th>cumulative_amount_clipped</th>\n",
       "      <th>cumulative_amount_clipped_and_normalized</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>29454</td>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>66056</td>\n",
       "      <td>2158-12-29 01:30:00</td>\n",
       "      <td>2158-12-29 01:31:00</td>\n",
       "      <td>2158-12-29 03:00:00</td>\n",
       "      <td>221468</td>\n",
       "      <td>10.000001</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.00</td>\n",
       "      <td>Diltiazem</td>\n",
       "      <td>27.966667</td>\n",
       "      <td>10.000001</td>\n",
       "      <td>521.454856</td>\n",
       "      <td>10.000001</td>\n",
       "      <td>0.019177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>29460</td>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>73391</td>\n",
       "      <td>2158-12-28 00:20:00</td>\n",
       "      <td>2158-12-28 00:21:00</td>\n",
       "      <td>2158-12-28 00:20:00</td>\n",
       "      <td>225975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>29444</td>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>32605</td>\n",
       "      <td>2158-12-28 10:44:00</td>\n",
       "      <td>2158-12-28 10:45:00</td>\n",
       "      <td>2158-12-28 10:44:00</td>\n",
       "      <td>225975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "      <td>13.200000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>29453</td>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>66056</td>\n",
       "      <td>2158-12-29 00:03:00</td>\n",
       "      <td>2158-12-29 00:04:00</td>\n",
       "      <td>2158-12-29 00:03:00</td>\n",
       "      <td>225975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "      <td>26.516667</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29461</td>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>82842</td>\n",
       "      <td>2158-12-28 11:22:00</td>\n",
       "      <td>2158-12-28 11:23:00</td>\n",
       "      <td>2158-12-28 11:22:00</td>\n",
       "      <td>221385</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Lorazepam (Ativan)</td>\n",
       "      <td>13.833333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>12.000001</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.083333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268029</th>\n",
       "      <td>167318</td>\n",
       "      <td>39997955</td>\n",
       "      <td>16185669</td>\n",
       "      <td>24274681</td>\n",
       "      <td>73391</td>\n",
       "      <td>2160-10-28 18:38:00</td>\n",
       "      <td>2160-10-29 03:20:00</td>\n",
       "      <td>2160-10-28 21:23:00</td>\n",
       "      <td>225152</td>\n",
       "      <td>9135.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>Stopped</td>\n",
       "      <td>25000.001953</td>\n",
       "      <td>1050.00</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>47.582778</td>\n",
       "      <td>59135.002693</td>\n",
       "      <td>83611.134865</td>\n",
       "      <td>59135.002693</td>\n",
       "      <td>0.707262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268030</th>\n",
       "      <td>167311</td>\n",
       "      <td>39997955</td>\n",
       "      <td>16185669</td>\n",
       "      <td>24274681</td>\n",
       "      <td>30753</td>\n",
       "      <td>2160-10-27 20:28:00</td>\n",
       "      <td>2160-10-27 20:29:00</td>\n",
       "      <td>2160-10-28 02:29:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>25.416111</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>31.075002</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.032180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268031</th>\n",
       "      <td>167314</td>\n",
       "      <td>39997955</td>\n",
       "      <td>16185669</td>\n",
       "      <td>24274681</td>\n",
       "      <td>30753</td>\n",
       "      <td>2160-10-28 02:29:00</td>\n",
       "      <td>2160-10-28 02:30:00</td>\n",
       "      <td>2160-10-28 02:29:00</td>\n",
       "      <td>225154</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>31.432778</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>31.075002</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.064360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268032</th>\n",
       "      <td>167317</td>\n",
       "      <td>39997955</td>\n",
       "      <td>16185669</td>\n",
       "      <td>24274681</td>\n",
       "      <td>62920</td>\n",
       "      <td>2160-10-28 10:03:00</td>\n",
       "      <td>2160-10-28 15:03:00</td>\n",
       "      <td>2160-10-28 14:04:00</td>\n",
       "      <td>225835</td>\n",
       "      <td>15.000001</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>0.05</td>\n",
       "      <td>Na Phos</td>\n",
       "      <td>38.999444</td>\n",
       "      <td>15.000001</td>\n",
       "      <td>60.000001</td>\n",
       "      <td>15.000001</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268033</th>\n",
       "      <td>167312</td>\n",
       "      <td>39997955</td>\n",
       "      <td>16185669</td>\n",
       "      <td>24274681</td>\n",
       "      <td>30753</td>\n",
       "      <td>2160-10-27 21:05:00</td>\n",
       "      <td>2160-10-27 21:06:00</td>\n",
       "      <td>2160-10-27 21:05:00</td>\n",
       "      <td>225910</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>FinishedRunning</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00</td>\n",
       "      <td>Pantoprazole (Protonix)</td>\n",
       "      <td>26.032778</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>319.636017</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.003129</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>268034 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         index   stay_id  subject_id   hadm_id  caregiver_id  \\\n",
       "0        29454  30004018    11101925  23920532         66056   \n",
       "1        29460  30004018    11101925  23920532         73391   \n",
       "2        29444  30004018    11101925  23920532         32605   \n",
       "3        29453  30004018    11101925  23920532         66056   \n",
       "4        29461  30004018    11101925  23920532         82842   \n",
       "...        ...       ...         ...       ...           ...   \n",
       "268029  167318  39997955    16185669  24274681         73391   \n",
       "268030  167311  39997955    16185669  24274681         30753   \n",
       "268031  167314  39997955    16185669  24274681         30753   \n",
       "268032  167317  39997955    16185669  24274681         62920   \n",
       "268033  167312  39997955    16185669  24274681         30753   \n",
       "\n",
       "                  starttime              endtime            storetime  itemid  \\\n",
       "0       2158-12-29 01:30:00  2158-12-29 01:31:00  2158-12-29 03:00:00  221468   \n",
       "1       2158-12-28 00:20:00  2158-12-28 00:21:00  2158-12-28 00:20:00  225975   \n",
       "2       2158-12-28 10:44:00  2158-12-28 10:45:00  2158-12-28 10:44:00  225975   \n",
       "3       2158-12-29 00:03:00  2158-12-29 00:04:00  2158-12-29 00:03:00  225975   \n",
       "4       2158-12-28 11:22:00  2158-12-28 11:23:00  2158-12-28 11:22:00  221385   \n",
       "...                     ...                  ...                  ...     ...   \n",
       "268029  2160-10-28 18:38:00  2160-10-29 03:20:00  2160-10-28 21:23:00  225152   \n",
       "268030  2160-10-27 20:28:00  2160-10-27 20:29:00  2160-10-28 02:29:00  225154   \n",
       "268031  2160-10-28 02:29:00  2160-10-28 02:30:00  2160-10-28 02:29:00  225154   \n",
       "268032  2160-10-28 10:03:00  2160-10-28 15:03:00  2160-10-28 14:04:00  225835   \n",
       "268033  2160-10-27 21:05:00  2160-10-27 21:06:00  2160-10-27 21:05:00  225910   \n",
       "\n",
       "             amount  ... continueinnextdept  statusdescription originalamount  \\\n",
       "0         10.000001  ...                  0    FinishedRunning      10.000000   \n",
       "1          1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "2          1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "3          1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "4          1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "...             ...  ...                ...                ...            ...   \n",
       "268029  9135.000000  ...                  0            Stopped   25000.001953   \n",
       "268030     1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "268031     1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "268032    15.000001  ...                  0    FinishedRunning      15.000000   \n",
       "268033     1.000000  ...                  0    FinishedRunning       1.000000   \n",
       "\n",
       "        originalrate                         label time_since_intime  \\\n",
       "0              10.00                     Diltiazem         27.966667   \n",
       "1               1.00  Heparin Sodium (Prophylaxis)          2.800000   \n",
       "2               1.00  Heparin Sodium (Prophylaxis)         13.200000   \n",
       "3               1.00  Heparin Sodium (Prophylaxis)         26.516667   \n",
       "4               1.00            Lorazepam (Ativan)         13.833333   \n",
       "...              ...                           ...               ...   \n",
       "268029       1050.00                Heparin Sodium         47.582778   \n",
       "268030          0.00              Morphine Sulfate         25.416111   \n",
       "268031          1.00              Morphine Sulfate         31.432778   \n",
       "268032          0.05                       Na Phos         38.999444   \n",
       "268033          1.00       Pantoprazole (Protonix)         26.032778   \n",
       "\n",
       "       cumulative_amount percentile_95 cumulative_amount_clipped  \\\n",
       "0              10.000001    521.454856                 10.000001   \n",
       "1               1.000000      6.000000                  1.000000   \n",
       "2               2.000000      6.000000                  2.000000   \n",
       "3               3.000000      6.000000                  3.000000   \n",
       "4               1.000000     12.000001                  1.000000   \n",
       "...                  ...           ...                       ...   \n",
       "268029      59135.002693  83611.134865              59135.002693   \n",
       "268030          1.000000     31.075002                  1.000000   \n",
       "268031          2.000000     31.075002                  2.000000   \n",
       "268032         15.000001     60.000001                 15.000001   \n",
       "268033          1.000000    319.636017                  1.000000   \n",
       "\n",
       "        cumulative_amount_clipped_and_normalized  \n",
       "0                                       0.019177  \n",
       "1                                       0.166667  \n",
       "2                                       0.333333  \n",
       "3                                       0.500000  \n",
       "4                                       0.083333  \n",
       "...                                          ...  \n",
       "268029                                  0.707262  \n",
       "268030                                  0.032180  \n",
       "268031                                  0.064360  \n",
       "268032                                  0.250000  \n",
       "268033                                  0.003129  \n",
       "\n",
       "[268034 rows x 33 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = meds_within_48h.reset_index()\n",
    "# sort per stay_id (patient) and medication (label) by time_sine_intime\n",
    "df_sorted = df.sort_values(by=['stay_id', 'label', 'time_since_intime'], ascending=True)\n",
    "# cumulative sum for each medication\n",
    "df_sorted['cumulative_amount'] = df_sorted.groupby(['stay_id', 'label'])['amount'].cumsum()\n",
    "# max cumulative sum per medication\n",
    "df_max_cumulative = df_sorted.groupby(['stay_id', 'label'])['cumulative_amount'].max().reset_index()\n",
    "# 95th percentile per medicaion\n",
    "percentile_95 = df_max_cumulative.groupby('label')['cumulative_amount'].quantile(0.95).reset_index()\n",
    "percentile_95.rename(columns={'cumulative_amount': 'percentile_95'}, inplace=True)\n",
    "# merge on medication\n",
    "df_sorted = df_sorted.merge(percentile_95, on='label', how='left')\n",
    "# clip to 95th percentile per medicaiton\n",
    "df_sorted['cumulative_amount_clipped'] = df_sorted['cumulative_amount'].where(\n",
    "    df_sorted['cumulative_amount'] <= df_sorted['percentile_95'],\n",
    "    df_sorted['percentile_95']\n",
    ")\n",
    "# normalize to make different medications (e.g. soluted Fetanyl and concentrated fetanyl) comparable in the same plot\n",
    "df_sorted['cumulative_amount_clipped_and_normalized'] = df_sorted['cumulative_amount_clipped'] / df_sorted['percentile_95']\n",
    "df_sorted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "98f347f0",
   "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>stay_id</th>\n",
       "      <th>label</th>\n",
       "      <th>cumulative_amount_clipped_and_normalized</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Diltiazem</td>\n",
       "      <td>0.019177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Lorazepam (Ativan)</td>\n",
       "      <td>0.291667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Metoprolol</td>\n",
       "      <td>0.181818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>0.096541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57485</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Acetaminophen-IV</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57486</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>0.707262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57487</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>0.064360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57488</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Na Phos</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57489</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Pantoprazole (Protonix)</td>\n",
       "      <td>0.003129</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>57490 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        stay_id                         label  \\\n",
       "0      30004018                     Diltiazem   \n",
       "1      30004018  Heparin Sodium (Prophylaxis)   \n",
       "2      30004018            Lorazepam (Ativan)   \n",
       "3      30004018                    Metoprolol   \n",
       "4      30004018              Morphine Sulfate   \n",
       "...         ...                           ...   \n",
       "57485  39997955              Acetaminophen-IV   \n",
       "57486  39997955                Heparin Sodium   \n",
       "57487  39997955              Morphine Sulfate   \n",
       "57488  39997955                       Na Phos   \n",
       "57489  39997955       Pantoprazole (Protonix)   \n",
       "\n",
       "       cumulative_amount_clipped_and_normalized  \n",
       "0                                      0.019177  \n",
       "1                                      0.500000  \n",
       "2                                      0.291667  \n",
       "3                                      0.181818  \n",
       "4                                      0.096541  \n",
       "...                                         ...  \n",
       "57485                                  0.333333  \n",
       "57486                                  0.707262  \n",
       "57487                                  0.064360  \n",
       "57488                                  0.250000  \n",
       "57489                                  0.003129  \n",
       "\n",
       "[57490 rows x 3 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_max_cumulative = df_sorted.groupby(['stay_id', 'label'])['cumulative_amount_clipped_and_normalized'].max().reset_index()\n",
    "df_max_cumulative"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "38db1228",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1200 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "top_labels = df_max_cumulative['label'].value_counts().nlargest(9).index\n",
    "df_subset = df_max_cumulative[df_max_cumulative['label'].isin(top_labels)]\n",
    "g = sns.FacetGrid(df_subset, col=\"label\", col_wrap=3, sharex=False, sharey=False, height=4)\n",
    "g.map(sns.histplot, \"cumulative_amount_clipped_and_normalized\", bins=30, kde=False)\n",
    "g.set_titles(\"{col_name}\")\n",
    "g.set_axis_labels(\"Cumulative Amount\", \"Frequency\")\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "184be6a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# stay_ids = pd.read_csv('data/meta.csv').stay_id\n",
    "# stay_ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "90adb150",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cp = sns.color_palette()\n",
    "markers = ['P','o','v','x','8','s','X','H','p','D']\n",
    "stay_ids = all_subjects.stay_id\n",
    "for stay_id in tqdm(stay_ids): # filtered_stay_ids\n",
    "    split = all_subjects[all_subjects.stay_id==stay_id].split.iloc[0]\n",
    "    if Path(f'data/meds2/{split}/{stay_id}.png').exists():\n",
    "        continue\n",
    "# stay_id = 31582758\n",
    "# fig,axs = plt.subplots(5,5,figsize=(10,10))\n",
    "    fig,axs = plt.subplots(3,4,figsize=(10,10))\n",
    "    axs = axs.flatten()\n",
    "    patient_meds = df_sorted[df_sorted.stay_id==stay_id]\n",
    "    for label in patient_meds.label.unique():\n",
    "        meds = patient_meds[patient_meds.label==label]\n",
    "        x = meds.time_since_intime\n",
    "        y = meds.cumulative_amount_clipped_and_normalized\n",
    "        # parent_category = med_category_assignment[label]\n",
    "        if label not in med_categories_t:\n",
    "            continue\n",
    "        parent_category = med_categories_t[label]\n",
    "        ax = axs[np.argwhere(np.array(list(med_categories.keys())) == parent_category)[0,0]]\n",
    "        # i = np.argwhere(np.array(med_category_assignment_t[parent_category])==label)[0,0]\n",
    "        i = np.argwhere(np.array(med_categories[parent_category])==label)[0,0]\n",
    "        ax.plot(x,y,color=cp[i],marker=markers[i])\n",
    "        ax.set_ylim([0,1])\n",
    "        ax.set_xlim([0,48])\n",
    "    for ax in axs:\n",
    "        ax.set_xticks([])\n",
    "        ax.set_yticks([])\n",
    "    fig.subplots_adjust(hspace=0., wspace=0.)\n",
    "    fig.savefig(f'data/meds2/{split}/{stay_id}.png', bbox_inches='tight', dpi=72)\n",
    "    plt.close(fig)\n",
    "    # break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "67d95426",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fig.savefig('images/med31582758.pdf',bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "6c24f866",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False, False, False, False, False, False, False,  True, False,\n",
       "       False])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(list(med_categories.keys())) == parent_category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3d8817d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(dict_keys(['Antiarrhythmics', 'Electrolyte Supplements', 'Beta-Blockers', 'Diuretics', 'Cardiac Glycosides', 'Insulins', 'Opioid Analgesics', 'Sedatives and Anesthetics', 'Thrombolytics', 'Vasopressors and Inotropes']),\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(med_categories.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "d90680de",
   "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>stay_id</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>time_since_intime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30004018</td>\n",
       "      <td>71.3</td>\n",
       "      <td>2.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30004018</td>\n",
       "      <td>71.3</td>\n",
       "      <td>13.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>30004018</td>\n",
       "      <td>71.3</td>\n",
       "      <td>13.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>30004018</td>\n",
       "      <td>71.3</td>\n",
       "      <td>15.966667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30004018</td>\n",
       "      <td>71.3</td>\n",
       "      <td>26.516667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268031</th>\n",
       "      <td>39997955</td>\n",
       "      <td>94.7</td>\n",
       "      <td>31.432778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268022</th>\n",
       "      <td>39997955</td>\n",
       "      <td>94.7</td>\n",
       "      <td>33.949444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268032</th>\n",
       "      <td>39997955</td>\n",
       "      <td>94.7</td>\n",
       "      <td>38.999444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268023</th>\n",
       "      <td>39997955</td>\n",
       "      <td>94.7</td>\n",
       "      <td>44.032778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268029</th>\n",
       "      <td>39997955</td>\n",
       "      <td>94.7</td>\n",
       "      <td>47.582778</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>268034 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         stay_id  patientweight  time_since_intime\n",
       "1       30004018           71.3           2.800000\n",
       "2       30004018           71.3          13.200000\n",
       "4       30004018           71.3          13.833333\n",
       "5       30004018           71.3          15.966667\n",
       "3       30004018           71.3          26.516667\n",
       "...          ...            ...                ...\n",
       "268031  39997955           94.7          31.432778\n",
       "268022  39997955           94.7          33.949444\n",
       "268032  39997955           94.7          38.999444\n",
       "268023  39997955           94.7          44.032778\n",
       "268029  39997955           94.7          47.582778\n",
       "\n",
       "[268034 rows x 3 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "patientweight = df_sorted[['stay_id','patientweight','time_since_intime']].sort_values(by=['stay_id', 'time_since_intime'], ascending=True)#.std().value_counts()\n",
    "patientweight.set_index('stay_id').to_csv('data/patientweight.csv')\n",
    "patientweight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "d4fa83bf-903e-41d3-8025-4b506a864872",
   "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>stay_id</th>\n",
       "      <th>label</th>\n",
       "      <th>max_cumulative_amount</th>\n",
       "      <th>min_time_since_intime</th>\n",
       "      <th>max_time_since_intime</th>\n",
       "      <th>prescription_count</th>\n",
       "      <th>delta_time</th>\n",
       "      <th>average_dosis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Diltiazem</td>\n",
       "      <td>10.000001</td>\n",
       "      <td>27.966667</td>\n",
       "      <td>27.966667</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Heparin Sodium (Prophylaxis)</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>26.516667</td>\n",
       "      <td>3</td>\n",
       "      <td>23.716667</td>\n",
       "      <td>0.126493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Lorazepam (Ativan)</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>13.833333</td>\n",
       "      <td>45.566667</td>\n",
       "      <td>5</td>\n",
       "      <td>31.733333</td>\n",
       "      <td>0.110294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Metoprolol</td>\n",
       "      <td>10.000001</td>\n",
       "      <td>28.466667</td>\n",
       "      <td>28.633333</td>\n",
       "      <td>2</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>60.000004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>30004018</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>34.966667</td>\n",
       "      <td>39.083333</td>\n",
       "      <td>2</td>\n",
       "      <td>4.116667</td>\n",
       "      <td>0.728745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57485</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Acetaminophen-IV</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>33.949444</td>\n",
       "      <td>44.032778</td>\n",
       "      <td>2</td>\n",
       "      <td>10.083333</td>\n",
       "      <td>198.347107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57486</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Heparin Sodium</td>\n",
       "      <td>59135.002693</td>\n",
       "      <td>1.449444</td>\n",
       "      <td>47.582778</td>\n",
       "      <td>6</td>\n",
       "      <td>46.133333</td>\n",
       "      <td>1281.828093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57487</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Morphine Sulfate</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>25.416111</td>\n",
       "      <td>31.432778</td>\n",
       "      <td>2</td>\n",
       "      <td>6.016667</td>\n",
       "      <td>0.332410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57488</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Na Phos</td>\n",
       "      <td>15.000001</td>\n",
       "      <td>38.999444</td>\n",
       "      <td>38.999444</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15.000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57489</th>\n",
       "      <td>39997955</td>\n",
       "      <td>Pantoprazole (Protonix)</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>26.032778</td>\n",
       "      <td>26.032778</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>57490 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        stay_id                         label  max_cumulative_amount  \\\n",
       "0      30004018                     Diltiazem              10.000001   \n",
       "1      30004018  Heparin Sodium (Prophylaxis)               3.000000   \n",
       "2      30004018            Lorazepam (Ativan)               3.500000   \n",
       "3      30004018                    Metoprolol              10.000001   \n",
       "4      30004018              Morphine Sulfate               3.000000   \n",
       "...         ...                           ...                    ...   \n",
       "57485  39997955              Acetaminophen-IV            2000.000000   \n",
       "57486  39997955                Heparin Sodium           59135.002693   \n",
       "57487  39997955              Morphine Sulfate               2.000000   \n",
       "57488  39997955                       Na Phos              15.000001   \n",
       "57489  39997955       Pantoprazole (Protonix)               1.000000   \n",
       "\n",
       "       min_time_since_intime  max_time_since_intime  prescription_count  \\\n",
       "0                  27.966667              27.966667                   1   \n",
       "1                   2.800000              26.516667                   3   \n",
       "2                  13.833333              45.566667                   5   \n",
       "3                  28.466667              28.633333                   2   \n",
       "4                  34.966667              39.083333                   2   \n",
       "...                      ...                    ...                 ...   \n",
       "57485              33.949444              44.032778                   2   \n",
       "57486               1.449444              47.582778                   6   \n",
       "57487              25.416111              31.432778                   2   \n",
       "57488              38.999444              38.999444                   1   \n",
       "57489              26.032778              26.032778                   1   \n",
       "\n",
       "       delta_time  average_dosis  \n",
       "0             NaN      10.000001  \n",
       "1       23.716667       0.126493  \n",
       "2       31.733333       0.110294  \n",
       "3        0.166667      60.000004  \n",
       "4        4.116667       0.728745  \n",
       "...           ...            ...  \n",
       "57485   10.083333     198.347107  \n",
       "57486   46.133333    1281.828093  \n",
       "57487    6.016667       0.332410  \n",
       "57488         NaN      15.000001  \n",
       "57489         NaN       1.000000  \n",
       "\n",
       "[57490 rows x 8 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_avg_dosis = df_sorted.groupby(['stay_id', 'label']).agg({\n",
    "    'cumulative_amount': 'max',\n",
    "    'time_since_intime': ['min', 'max'],\n",
    "    'amount': 'count'\n",
    "}).reset_index()\n",
    "df_avg_dosis.columns = ['stay_id', 'label', 'max_cumulative_amount', 'min_time_since_intime', 'max_time_since_intime','prescription_count']\n",
    "df_avg_dosis['delta_time'] = df_avg_dosis['max_time_since_intime'] - df_avg_dosis['min_time_since_intime']\n",
    "df_avg_dosis['delta_time'] = df_avg_dosis['delta_time'].replace(0, np.nan)\n",
    "df_avg_dosis['average_dosis'] = df_avg_dosis['max_cumulative_amount'] / df_avg_dosis['delta_time']\n",
    "df_avg_dosis['average_dosis'] = df_avg_dosis['average_dosis'].fillna(df_avg_dosis['max_cumulative_amount'])\n",
    "df_avg_dosis.set_index('stay_id').to_csv('data/medication_doses.csv')\n",
    "df_avg_dosis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "cf959e3b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "stay_id                                          30004018\n",
       "subject_id                                       11101925\n",
       "hadm_id                                          23920532\n",
       "caregiver_id                                        66056\n",
       "starttime                             2158-12-29 03:20:00\n",
       "endtime                               2158-12-29 04:15:00\n",
       "storetime                             2158-12-29 04:12:00\n",
       "itemid                                             221906\n",
       "amount                                           0.117647\n",
       "amountuom                                              mg\n",
       "rate                                             0.030001\n",
       "rateuom                                        mcg/kg/min\n",
       "orderid                                           1866448\n",
       "linkorderid                                       1866448\n",
       "ordercategoryname                                01-Drips\n",
       "secondaryordercategoryname       02-Fluids (Crystalloids)\n",
       "ordercomponenttypedescription        Main order parameter\n",
       "ordercategorydescription                   Continuous Med\n",
       "patientweight                                        71.3\n",
       "totalamount                                         250.0\n",
       "totalamountuom                                         ml\n",
       "isopenbag                                               0\n",
       "continueinnextdept                                      0\n",
       "statusdescription                         ChangeDose/Rate\n",
       "originalamount                                        8.0\n",
       "originalrate                                         0.03\n",
       "label                                      Norepinephrine\n",
       "time_since_intime                                    29.8\n",
       "Name: 29457, dtype: object"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = meds_within_48h.reset_index()\n",
    "# sort per stay_id (patient) and medication (label) by time_sine_intime\n",
    "df_sorted = df.sort_values(by=['stay_id', 'label', 'time_since_intime'], ascending=True)\n",
    "df_sorted.dropna(subset=['rate']).iloc[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "42461e98",
   "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>stay_id</th>\n",
       "      <th>subject_id</th>\n",
       "      <th>hadm_id</th>\n",
       "      <th>caregiver_id</th>\n",
       "      <th>starttime</th>\n",
       "      <th>endtime</th>\n",
       "      <th>storetime</th>\n",
       "      <th>itemid</th>\n",
       "      <th>amount</th>\n",
       "      <th>amountuom</th>\n",
       "      <th>...</th>\n",
       "      <th>patientweight</th>\n",
       "      <th>totalamount</th>\n",
       "      <th>totalamountuom</th>\n",
       "      <th>isopenbag</th>\n",
       "      <th>continueinnextdept</th>\n",
       "      <th>statusdescription</th>\n",
       "      <th>originalamount</th>\n",
       "      <th>originalrate</th>\n",
       "      <th>label</th>\n",
       "      <th>time_since_intime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>29457</th>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>66056</td>\n",
       "      <td>2158-12-29 03:20:00</td>\n",
       "      <td>2158-12-29 04:15:00</td>\n",
       "      <td>2158-12-29 04:12:00</td>\n",
       "      <td>221906</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>71.3</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.03</td>\n",
       "      <td>Norepinephrine</td>\n",
       "      <td>29.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29458</th>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>66056</td>\n",
       "      <td>2158-12-29 04:15:00</td>\n",
       "      <td>2158-12-29 05:10:00</td>\n",
       "      <td>2158-12-29 06:01:00</td>\n",
       "      <td>221906</td>\n",
       "      <td>0.235230</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>71.3</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>ChangeDose/Rate</td>\n",
       "      <td>7.882353</td>\n",
       "      <td>0.06</td>\n",
       "      <td>Norepinephrine</td>\n",
       "      <td>30.716667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29459</th>\n",
       "      <td>30004018</td>\n",
       "      <td>11101925</td>\n",
       "      <td>23920532</td>\n",
       "      <td>66056</td>\n",
       "      <td>2158-12-29 05:10:00</td>\n",
       "      <td>2158-12-29 08:36:00</td>\n",
       "      <td>2158-12-29 06:01:00</td>\n",
       "      <td>221906</td>\n",
       "      <td>1.174726</td>\n",
       "      <td>mg</td>\n",
       "      <td>...</td>\n",
       "      <td>71.3</td>\n",
       "      <td>250.0</td>\n",
       "      <td>ml</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Stopped</td>\n",
       "      <td>7.647122</td>\n",
       "      <td>0.08</td>\n",
       "      <td>Norepinephrine</td>\n",
       "      <td>31.633333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        stay_id  subject_id   hadm_id  caregiver_id           starttime  \\\n",
       "29457  30004018    11101925  23920532         66056 2158-12-29 03:20:00   \n",
       "29458  30004018    11101925  23920532         66056 2158-12-29 04:15:00   \n",
       "29459  30004018    11101925  23920532         66056 2158-12-29 05:10:00   \n",
       "\n",
       "                   endtime            storetime  itemid    amount amountuom  \\\n",
       "29457  2158-12-29 04:15:00  2158-12-29 04:12:00  221906  0.117647        mg   \n",
       "29458  2158-12-29 05:10:00  2158-12-29 06:01:00  221906  0.235230        mg   \n",
       "29459  2158-12-29 08:36:00  2158-12-29 06:01:00  221906  1.174726        mg   \n",
       "\n",
       "       ...  patientweight totalamount  totalamountuom  isopenbag  \\\n",
       "29457  ...           71.3       250.0              ml          0   \n",
       "29458  ...           71.3       250.0              ml          0   \n",
       "29459  ...           71.3       250.0              ml          0   \n",
       "\n",
       "      continueinnextdept statusdescription originalamount originalrate  \\\n",
       "29457                  0   ChangeDose/Rate       8.000000         0.03   \n",
       "29458                  0   ChangeDose/Rate       7.882353         0.06   \n",
       "29459                  0           Stopped       7.647122         0.08   \n",
       "\n",
       "                label  time_since_intime  \n",
       "29457  Norepinephrine          29.800000  \n",
       "29458  Norepinephrine          30.716667  \n",
       "29459  Norepinephrine          31.633333  \n",
       "\n",
       "[3 rows x 28 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sorted[(df_sorted.stay_id==30004018)&(df_sorted.label=='Norepinephrine')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0305bfe5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['FinishedRunning', 'ChangeDose/Rate', 'Stopped', 'Paused', 'Bolus'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sorted.statusdescription.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "93f7ae6b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "cfm",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}