notebooks/1-Table-Baseline-Characteristics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ICAN clinical data characteristics\n",
    "\n",
    "**Objective:** present the simulated ICAN database."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Packages importation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy import stats\n",
    "\n",
    "import Missing_Values as MV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2505, 28)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ican = pd.read_csv(\"simulated_ican_clinical_data.csv\", sep = \"#\")\n",
    "ican.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Baseline Characteristics table creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_UIA = ican[ican.rupture == \"No\"]\n",
    "data_RIA = ican[ican.rupture == \"Yes\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f_dec(x) :\n",
    "    if x >= 10 :\n",
    "        return(1)\n",
    "    else : \n",
    "        return(2) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def info_count(df, var, value, dec = 1) :\n",
    "    count = sum(df[var] == value)\n",
    "    if count == 0 :\n",
    "        return(0)\n",
    "    else :\n",
    "        return(str(count) + \" (\" + format(count/len(df)*100, '0.1f') + \"%)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def info_mean(df, var, nonnormal_vect = None) :\n",
    "    if (var in nonnormal_vect) :\n",
    "        dec = f_dec(np.median(df[var]))\n",
    "        return(format(np.median(df[var]), '0.'+str(dec)+'f') + \" [\" \n",
    "               + format(np.quantile(df[var], 0.25), '0.'+str(dec)+'f') + \"-\" \n",
    "               + format(np.quantile(df[var], 0.75), '0.'+str(dec)+'f') + \"]\")\n",
    "    else :\n",
    "        dec = f_dec(np.mean(df[var]))\n",
    "        return(format(np.mean(df[var]), '0.'+str(dec)+'f') + \" \\u00B1 \"\n",
    "               + format(np.std(df[var]), '0.'+str(dec)+'f'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def col_filling_table(df, nonnormal_vect = None) :\n",
    "    col = []\n",
    "    table_index = []\n",
    "    col.append(info_count(df, \"sex\", \"F\")) ; table_index.append(\"Sex (women)\")\n",
    "    col.append(info_mean(df, \"age\", nonnormal_vect)) ; table_index.append(\"Age (year)\")\n",
    "    col.append(info_mean(df, \"bmi\", nonnormal_vect)) ; table_index.append(\"Body mass index (kg/m²)\")\n",
    "    col.append(info_count(df, \"sporadic_case\", \"No\")) ; table_index.append(\"Familial history (yes)\")\n",
    "    col.append(info_mean(df, \"adjusted_size_ratio\", nonnormal_vect)) ; table_index.append(\"Adjusted Size Ratio\")\n",
    "    col.append(info_count(df, \"IA_location\", \"ICA\")) ; table_index.append(\"IA location - ICA\")\n",
    "    col.append(info_count(df, \"IA_location\", \"MCA\")) ; table_index.append(\"IA location - MCA\")\n",
    "    col.append(info_count(df, \"IA_location\", \"ACA\")) ; table_index.append(\"IA location - ACA\")\n",
    "    col.append(info_count(df, \"IA_location\", \"PCA\")) ; table_index.append(\"IA location - PCA\")\n",
    "    col.append(info_count(df, \"multiple_IA\", \"Yes\")) ; table_index.append(\"Multiple IA (yes)\")\n",
    "    col.append(info_count(df, \"hta\", \"Yes\")) ; table_index.append(\"Hypertension (yes)\")\n",
    "    col.append(info_count(df, \"headaches\", \"Yes\")) ; table_index.append(\"Headaches (yes)\")\n",
    "    col.append(info_count(df, \"dyslipidemia\", \"Yes\")) ; table_index.append(\"Dyslipidemia (yes)\")\n",
    "    col.append(info_count(df, \"ischemic_stroke_history\", \"Yes\")) ; table_index.append(\"Ischemic stroke history (yes)\")\n",
    "    \n",
    "    col.append(info_count(df, \"ischemic_heart_disease_history\", \"Yes\"))\n",
    "    table_index.append(\"Ischemic heart disease history (yes)\")\n",
    "    \n",
    "    col.append(info_count(df, \"statin_ttt\", \"Yes\")) ; table_index.append(\"Statin treatment (yes)\")\n",
    "    col.append(info_count(df, \"platelet_aggregation_inhibiting_ttt\", \"Yes\")) ; \n",
    "    table_index.append(\"Platelet aggregation inhibiting treatment (yes)\")\n",
    "    col.append(info_count(df, \"hormone_therapy_ttt\", \"Yes\")) ; table_index.append(\"Hormone therapy treatment (yes)\")\n",
    "    col.append(info_count(df, \"allergy\", \"Yes\")) ; table_index.append(\"Allergy (yes)\")\n",
    "    \n",
    "    col.append(info_count(df, \"tobacco\", \"Non-smoker\"))\n",
    "    table_index.append(\"Tobacco consumption - Non-smoker\")\n",
    "    \n",
    "    col.append(info_count(df, \"tobacco\", \"Minor_smoker\"))\n",
    "    table_index.append(\"Tobacco consumption - Minor smoker\")\n",
    "    \n",
    "    col.append(info_count(df, \"tobacco\", \"Regular_smoker\"))\n",
    "    table_index.append(\"Tobacco consumption - Regular smoker\")\n",
    "    \n",
    "    col.append(info_count(df, \"alcohol\", \"> 150g\")) ; table_index.append(\"Alcohol consumption (> 150g)\")\n",
    "    return col, table_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "nonnormal_vect = [\"adjusted_size_ratio\"]\n",
    "col_UIA, table_index = col_filling_table(data_UIA, nonnormal_vect)\n",
    "col_RIA, table_index = col_filling_table(data_RIA, nonnormal_vect)\n",
    "col_TOT, table_index = col_filling_table(ican, nonnormal_vect)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "table = pd.DataFrame({\"All (n = \" + str(len(ican)) + ')' : col_TOT, \n",
    "                      \"RIA (n = \" + str(len(data_RIA)) + ')' : col_RIA, \n",
    "                      \"UIA (n = \" + str(len(data_UIA)) + ')' : col_UIA}, \n",
    "                     index = table_index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>All (n = 2505)</th>\n",
       "      <th>RIA (n = 959)</th>\n",
       "      <th>UIA (n = 1546)</th>\n",
       "      <th>p-value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Sex (women)</th>\n",
       "      <td>1784 (71.2%)</td>\n",
       "      <td>667 (69.6%)</td>\n",
       "      <td>1117 (72.3%)</td>\n",
       "      <td>0.160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Age (year)</th>\n",
       "      <td>53.8 ± 12.3</td>\n",
       "      <td>50.8 ± 11.8</td>\n",
       "      <td>55.7 ± 12.2</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Body mass index (kg/m²)</th>\n",
       "      <td>25.1 ± 4.5</td>\n",
       "      <td>24.9 ± 4.4</td>\n",
       "      <td>25.2 ± 4.6</td>\n",
       "      <td>0.101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Familial history (yes)</th>\n",
       "      <td>509 (20.3%)</td>\n",
       "      <td>169 (17.6%)</td>\n",
       "      <td>340 (22.0%)</td>\n",
       "      <td>0.010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adjusted Size Ratio</th>\n",
       "      <td>3.27 [1.94-5.51]</td>\n",
       "      <td>3.96 [2.27-6.12]</td>\n",
       "      <td>2.89 [1.79-5.05]</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IA location - ICA</th>\n",
       "      <td>610 (24.4%)</td>\n",
       "      <td>115 (12.0%)</td>\n",
       "      <td>495 (32.0%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IA location - MCA</th>\n",
       "      <td>666 (26.6%)</td>\n",
       "      <td>243 (25.3%)</td>\n",
       "      <td>423 (27.4%)</td>\n",
       "      <td>0.286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IA location - ACA</th>\n",
       "      <td>766 (30.6%)</td>\n",
       "      <td>374 (39.0%)</td>\n",
       "      <td>392 (25.4%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IA location - PCA</th>\n",
       "      <td>463 (18.5%)</td>\n",
       "      <td>227 (23.7%)</td>\n",
       "      <td>236 (15.3%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Multiple IA (yes)</th>\n",
       "      <td>802 (32.0%)</td>\n",
       "      <td>333 (34.7%)</td>\n",
       "      <td>469 (30.3%)</td>\n",
       "      <td>0.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hypertension (yes)</th>\n",
       "      <td>920 (36.7%)</td>\n",
       "      <td>342 (35.7%)</td>\n",
       "      <td>578 (37.4%)</td>\n",
       "      <td>0.408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Headaches (yes)</th>\n",
       "      <td>473 (18.9%)</td>\n",
       "      <td>166 (17.3%)</td>\n",
       "      <td>307 (19.9%)</td>\n",
       "      <td>0.126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dyslipidemia (yes)</th>\n",
       "      <td>567 (22.6%)</td>\n",
       "      <td>172 (17.9%)</td>\n",
       "      <td>395 (25.5%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ischemic stroke history (yes)</th>\n",
       "      <td>172 (6.9%)</td>\n",
       "      <td>23 (2.4%)</td>\n",
       "      <td>149 (9.6%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ischemic heart disease history (yes)</th>\n",
       "      <td>104 (4.2%)</td>\n",
       "      <td>18 (1.9%)</td>\n",
       "      <td>86 (5.6%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Statin treatment (yes)</th>\n",
       "      <td>272 (10.9%)</td>\n",
       "      <td>79 (8.2%)</td>\n",
       "      <td>193 (12.5%)</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Platelet aggregation inhibiting treatment (yes)</th>\n",
       "      <td>320 (12.8%)</td>\n",
       "      <td>90 (9.4%)</td>\n",
       "      <td>230 (14.9%)</td>\n",
       "      <td>&lt;0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hormone therapy treatment (yes)</th>\n",
       "      <td>74 (3.0%)</td>\n",
       "      <td>15 (1.6%)</td>\n",
       "      <td>59 (3.8%)</td>\n",
       "      <td>0.002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Allergy (yes)</th>\n",
       "      <td>568 (22.7%)</td>\n",
       "      <td>206 (21.5%)</td>\n",
       "      <td>362 (23.4%)</td>\n",
       "      <td>0.282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tobacco consumption - Non-smoker</th>\n",
       "      <td>738 (29.5%)</td>\n",
       "      <td>280 (29.2%)</td>\n",
       "      <td>458 (29.6%)</td>\n",
       "      <td>0.855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tobacco consumption - Minor smoker</th>\n",
       "      <td>1057 (42.2%)</td>\n",
       "      <td>425 (44.3%)</td>\n",
       "      <td>632 (40.9%)</td>\n",
       "      <td>0.099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tobacco consumption - Regular smoker</th>\n",
       "      <td>710 (28.3%)</td>\n",
       "      <td>254 (26.5%)</td>\n",
       "      <td>456 (29.5%)</td>\n",
       "      <td>0.114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alcohol consumption (&gt; 150g)</th>\n",
       "      <td>301 (12.0%)</td>\n",
       "      <td>127 (13.2%)</td>\n",
       "      <td>174 (11.3%)</td>\n",
       "      <td>0.154</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   All (n = 2505)  \\\n",
       "Sex (women)                                          1784 (71.2%)   \n",
       "Age (year)                                            53.8 ± 12.3   \n",
       "Body mass index (kg/m²)                                25.1 ± 4.5   \n",
       "Familial history (yes)                                509 (20.3%)   \n",
       "Adjusted Size Ratio                              3.27 [1.94-5.51]   \n",
       "IA location - ICA                                     610 (24.4%)   \n",
       "IA location - MCA                                     666 (26.6%)   \n",
       "IA location - ACA                                     766 (30.6%)   \n",
       "IA location - PCA                                     463 (18.5%)   \n",
       "Multiple IA (yes)                                     802 (32.0%)   \n",
       "Hypertension (yes)                                    920 (36.7%)   \n",
       "Headaches (yes)                                       473 (18.9%)   \n",
       "Dyslipidemia (yes)                                    567 (22.6%)   \n",
       "Ischemic stroke history (yes)                          172 (6.9%)   \n",
       "Ischemic heart disease history (yes)                   104 (4.2%)   \n",
       "Statin treatment (yes)                                272 (10.9%)   \n",
       "Platelet aggregation inhibiting treatment (yes)       320 (12.8%)   \n",
       "Hormone therapy treatment (yes)                         74 (3.0%)   \n",
       "Allergy (yes)                                         568 (22.7%)   \n",
       "Tobacco consumption - Non-smoker                      738 (29.5%)   \n",
       "Tobacco consumption - Minor smoker                   1057 (42.2%)   \n",
       "Tobacco consumption - Regular smoker                  710 (28.3%)   \n",
       "Alcohol consumption (> 150g)                          301 (12.0%)   \n",
       "\n",
       "                                                    RIA (n = 959)  \\\n",
       "Sex (women)                                           667 (69.6%)   \n",
       "Age (year)                                            50.8 ± 11.8   \n",
       "Body mass index (kg/m²)                                24.9 ± 4.4   \n",
       "Familial history (yes)                                169 (17.6%)   \n",
       "Adjusted Size Ratio                              3.96 [2.27-6.12]   \n",
       "IA location - ICA                                     115 (12.0%)   \n",
       "IA location - MCA                                     243 (25.3%)   \n",
       "IA location - ACA                                     374 (39.0%)   \n",
       "IA location - PCA                                     227 (23.7%)   \n",
       "Multiple IA (yes)                                     333 (34.7%)   \n",
       "Hypertension (yes)                                    342 (35.7%)   \n",
       "Headaches (yes)                                       166 (17.3%)   \n",
       "Dyslipidemia (yes)                                    172 (17.9%)   \n",
       "Ischemic stroke history (yes)                           23 (2.4%)   \n",
       "Ischemic heart disease history (yes)                    18 (1.9%)   \n",
       "Statin treatment (yes)                                  79 (8.2%)   \n",
       "Platelet aggregation inhibiting treatment (yes)         90 (9.4%)   \n",
       "Hormone therapy treatment (yes)                         15 (1.6%)   \n",
       "Allergy (yes)                                         206 (21.5%)   \n",
       "Tobacco consumption - Non-smoker                      280 (29.2%)   \n",
       "Tobacco consumption - Minor smoker                    425 (44.3%)   \n",
       "Tobacco consumption - Regular smoker                  254 (26.5%)   \n",
       "Alcohol consumption (> 150g)                          127 (13.2%)   \n",
       "\n",
       "                                                   UIA (n = 1546) p-value  \n",
       "Sex (women)                                          1117 (72.3%)   0.160  \n",
       "Age (year)                                            55.7 ± 12.2  <0.001  \n",
       "Body mass index (kg/m²)                                25.2 ± 4.6   0.101  \n",
       "Familial history (yes)                                340 (22.0%)   0.010  \n",
       "Adjusted Size Ratio                              2.89 [1.79-5.05]  <0.001  \n",
       "IA location - ICA                                     495 (32.0%)  <0.001  \n",
       "IA location - MCA                                     423 (27.4%)   0.286  \n",
       "IA location - ACA                                     392 (25.4%)  <0.001  \n",
       "IA location - PCA                                     236 (15.3%)  <0.001  \n",
       "Multiple IA (yes)                                     469 (30.3%)   0.025  \n",
       "Hypertension (yes)                                    578 (37.4%)   0.408  \n",
       "Headaches (yes)                                       307 (19.9%)   0.126  \n",
       "Dyslipidemia (yes)                                    395 (25.5%)  <0.001  \n",
       "Ischemic stroke history (yes)                          149 (9.6%)  <0.001  \n",
       "Ischemic heart disease history (yes)                    86 (5.6%)  <0.001  \n",
       "Statin treatment (yes)                                193 (12.5%)   0.001  \n",
       "Platelet aggregation inhibiting treatment (yes)       230 (14.9%)  <0.001  \n",
       "Hormone therapy treatment (yes)                         59 (3.8%)   0.002  \n",
       "Allergy (yes)                                         362 (23.4%)   0.282  \n",
       "Tobacco consumption - Non-smoker                      458 (29.6%)   0.855  \n",
       "Tobacco consumption - Minor smoker                    632 (40.9%)   0.099  \n",
       "Tobacco consumption - Regular smoker                  456 (29.5%)   0.114  \n",
       "Alcohol consumption (> 150g)                          174 (11.3%)   0.154  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Pvalues = []\n",
    "for i in range(len(table)) :\n",
    "    if \"%\" in table.iloc[i,1] :\n",
    "        val_Yes = [int(table.iloc[i,1].split()[0]), int(table.iloc[i,2].split()[0])]\n",
    "        val_No = [sum(ican.rupture == \"Yes\") - int(table.iloc[i,1].split()[0]), \n",
    "                  sum(ican.rupture == \"No\") - int(table.iloc[i,2].split()[0])]\n",
    "        df_contingence = pd.DataFrame({\"Yes\" : val_Yes, \"No\" : val_No})\n",
    "        st_chi2, st_p, st_dof, st_exp = stats.chi2_contingency(df_contingence)\n",
    "        if st_p < 0.001 :\n",
    "            Pvalues.append(\"<0.001\")\n",
    "        else : \n",
    "            Pvalues.append(format(st_p, '0.3f'))\n",
    "    elif \"Age\" in table.iloc[i].name :\n",
    "        st_ttest, st_p = stats.ttest_ind(data_RIA.age, data_UIA.age, axis = 0, equal_var = False)\n",
    "        if st_p < 0.001 :\n",
    "            Pvalues.append(\"<0.001\")\n",
    "        else : \n",
    "            Pvalues.append(format(st_p, '0.3f'))\n",
    "    elif \"Body\" in table.iloc[i].name :\n",
    "        st_ttest, st_p = stats.ttest_ind(data_RIA.bmi, data_UIA.bmi, axis = 0, equal_var = False)\n",
    "        if st_p < 0.001 :\n",
    "            Pvalues.append(\"<0.001\")\n",
    "        else : \n",
    "            Pvalues.append(format(st_p, '0.3f'))\n",
    "    elif \"Adjusted Size Ratio\" in table.iloc[i].name :\n",
    "        st_med, st_p, st_m, st_table = stats.median_test(data_RIA.adjusted_size_ratio, data_UIA.adjusted_size_ratio)\n",
    "        if st_p < 0.001 :\n",
    "            Pvalues.append(\"<0.001\")\n",
    "        else : \n",
    "            Pvalues.append(format(st_p, '0.3f'))\n",
    "table[\"p-value\"] = Pvalues\n",
    "table"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}