2.ecg.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```\n",
    "# Copyright 2023 by the author(s) of CHI2023 Late Breaking Work\n",
    "# \"Exploring Physiological Correlates of Visual Complexity Adaptation:\n",
    "# Insights from EDA, ECG, and EEG Data for Adaptation Evaluation in VR\n",
    "# Adaptive Systems\". All rights reserved.\n",
    "#\n",
    "# Use of this source code is governed by a GPLv3 license that\n",
    "# can be found in the LICENSE file.\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tqdm\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import neurokit2 as nk\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['text.usetex'] = True\n",
    "plt.rcParams['text.latex.preamble'] = r'''\\usepackage{libertine}\\usepackage{fontawesome5}\\usepackage[dvipsnames]{xcolor}'''\n",
    "plt.rcParams['ps.usedistiller'] = 'xpdf'\n",
    "import seaborn as sns\n",
    "sns.set_theme(style='ticks')\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "from lbw import get_event_intervals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_ecg(pid):\n",
    "\n",
    "    dfECG = pd.read_csv(f\"./data/ID{pid}-ECG.csv\")\n",
    "    dfState = pd.read_csv(f\"./data/ID{pid}-state.csv\")\n",
    "    dfAdaptation = pd.read_csv(f\"./data/ID{pid}-adaptation.csv\")\n",
    "    dfECG.drop(\"TimeLsl\", axis =1, inplace=True)\n",
    "\n",
    "    intervals, workloads = get_event_intervals(dfAdaptation)\n",
    "    for idx, interval in enumerate(intervals):\n",
    "        dfECG.loc[(dfECG['Time'] >= interval[0]) & (dfECG['Time'] <= interval[1]), 'workload'] = workloads[idx]\n",
    "        dfECG.loc[(dfECG['Time'] >= interval[0]) & (dfECG['Time'] <= interval[1]), 'interval_id'] = idx\n",
    "    dfECG = dfECG[dfECG['workload'].notna()]\n",
    "\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][2]) & (dfECG.Time < dfState['Time'][2+1])].index, 'BlockNumber']   = dfState['BlockNumber'][2]\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][4]) & (dfECG.Time < dfState['Time'][4+1])].index, 'BlockNumber']   = dfState['BlockNumber'][4]\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][6]) & (dfECG.Time < dfState['Time'][6+1])].index, 'BlockNumber']   = dfState['BlockNumber'][6]\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][8]) & (dfECG.Time < dfState['Time'][8+1])].index, 'BlockNumber']  = dfState['BlockNumber'][8]\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][10]) & (dfECG.Time < dfState['Time'][10+1])].index, 'BlockNumber'] = dfState['BlockNumber'][10]\n",
    "    dfECG.loc[dfECG[(dfECG.Time > dfState['Time'][12]) & (dfECG.Time < dfState['Time'][12+1])].index, 'BlockNumber'] = dfState['BlockNumber'][12]\n",
    "    dfECG = dfECG.dropna()\n",
    "    dfECG = dfECG.sort_values(by=\"Time\")\n",
    "    return dfECG\n",
    "\n",
    "def extract_ecg_features(pid, df):\n",
    "    df_features = pd.DataFrame()\n",
    "    intervals = df.interval_id.unique()\n",
    "    for interval_id in intervals:\n",
    "        df_current = df[df.interval_id == interval_id]\n",
    "\n",
    "        if len(df_current['Value']) < 230:\n",
    "            df_features = df_features.append(pd.DataFrame.from_dict([{\n",
    "                'pid': pid,\n",
    "                'workload': df_current['workload'].iloc[0],\n",
    "                'interval': interval_id,\n",
    "            }]))\n",
    "            continue\n",
    "\n",
    "        # # H10 Technical Specification\n",
    "        # # Sample rate = 130 Hz ± 2 % (Tamb = +20 … +40 °C)\n",
    "        # #               130 Hz ± 5 % (Tamb = -20 … +70 °C)\n",
    "        # # https://github.com/polarofficial/polar-ble-sdk/blob/master/technical_documentation/H10_ECG_Explained.docx\n",
    "        s, _ = nk.ecg_process(df_current['Value'], sampling_rate=130)\n",
    "        peaks, _ = nk.ecg_peaks(nk.ecg_clean(df_current['Value'], sampling_rate=130),\n",
    "                                sampling_rate=130,\n",
    "                                correct_artifacts=True)\n",
    "\n",
    "        try:\n",
    "            hrv_rmssd = np.mean(nk.hrv(peaks, sampling_rate=130)['HRV_RMSSD'])\n",
    "        except Exception:\n",
    "            hrv_rmssd = np.NaN\n",
    "\n",
    "\n",
    "        df_features = df_features.append(pd.DataFrame.from_dict([{\n",
    "            'pid': pid,\n",
    "            'workload': df_current['workload'].iloc[0],\n",
    "            'interval': interval_id,\n",
    "            # https://neuropsychology.github.io/NeuroKit/functions/ecg.html?highlight=ecg_rate\n",
    "            'ecg_rate_mean': np.mean(s['ECG_Rate']),\n",
    "            'ecg_rate_var': np.var(s['ECG_Rate']),\n",
    "            # https://neuropsychology.github.io/NeuroKit/examples/ecg_hrv/ecg_hrv.html?highlight=rmssd\n",
    "            'hrv_rmssd': hrv_rmssd,\n",
    "        }]))\n",
    "    return df_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 8/8 [00:01<00:00,  5.22it/s]\n"
     ]
    }
   ],
   "source": [
    "pid_valid = [4, 5, 7, 9, 10, 11, 12, 13]\n",
    "df_all = {}\n",
    "\n",
    "for pid in tqdm.tqdm(pid_valid):\n",
    "    df = load_ecg(pid)\n",
    "    df_all[pid] = df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>pid</th>\n",
       "      <th>workload</th>\n",
       "      <th>interval</th>\n",
       "      <th>ecg_rate_mean</th>\n",
       "      <th>ecg_rate_var</th>\n",
       "      <th>hrv_rmssd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>low</td>\n",
       "      <td>0.0</td>\n",
       "      <td>101.691743</td>\n",
       "      <td>768.889866</td>\n",
       "      <td>305.385272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>high</td>\n",
       "      <td>1.0</td>\n",
       "      <td>82.268868</td>\n",
       "      <td>561.024381</td>\n",
       "      <td>378.258504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>low</td>\n",
       "      <td>0.0</td>\n",
       "      <td>81.405105</td>\n",
       "      <td>666.763541</td>\n",
       "      <td>363.035634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>high</td>\n",
       "      <td>1.0</td>\n",
       "      <td>84.830229</td>\n",
       "      <td>720.261262</td>\n",
       "      <td>387.400175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>low</td>\n",
       "      <td>2.0</td>\n",
       "      <td>101.942719</td>\n",
       "      <td>1248.513926</td>\n",
       "      <td>317.491622</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>low</td>\n",
       "      <td>10.0</td>\n",
       "      <td>88.520034</td>\n",
       "      <td>903.015398</td>\n",
       "      <td>392.783097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>high</td>\n",
       "      <td>11.0</td>\n",
       "      <td>102.132568</td>\n",
       "      <td>1149.148648</td>\n",
       "      <td>320.594861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>low</td>\n",
       "      <td>12.0</td>\n",
       "      <td>95.792852</td>\n",
       "      <td>825.593806</td>\n",
       "      <td>239.975514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>high</td>\n",
       "      <td>13.0</td>\n",
       "      <td>113.298078</td>\n",
       "      <td>242.748618</td>\n",
       "      <td>72.813290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>low</td>\n",
       "      <td>14.0</td>\n",
       "      <td>85.453865</td>\n",
       "      <td>706.068897</td>\n",
       "      <td>328.785160</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>87 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    pid workload  interval  ecg_rate_mean  ecg_rate_var   hrv_rmssd\n",
       "0     4      low       0.0     101.691743    768.889866  305.385272\n",
       "0     4     high       1.0      82.268868    561.024381  378.258504\n",
       "0     5      low       0.0      81.405105    666.763541  363.035634\n",
       "0     5     high       1.0      84.830229    720.261262  387.400175\n",
       "0     5      low       2.0     101.942719   1248.513926  317.491622\n",
       "..  ...      ...       ...            ...           ...         ...\n",
       "0    13      low      10.0      88.520034    903.015398  392.783097\n",
       "0    13     high      11.0     102.132568   1149.148648  320.594861\n",
       "0    13      low      12.0      95.792852    825.593806  239.975514\n",
       "0    13     high      13.0     113.298078    242.748618   72.813290\n",
       "0    13      low      14.0      85.453865    706.068897  328.785160\n",
       "\n",
       "[87 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame()\n",
    "for pid in pid_valid:\n",
    "    df_features = extract_ecg_features(pid, df_all[pid])\n",
    "    df = df.append(df_features)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"./processed/ecg.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(6, 5))\n",
    "colors = list({'yellow' : '#FBC63D', 'blue': '#0376BB'}.values())\n",
    "sns.boxplot(x='workload', y='ecg_rate_mean', data=df, ax=axes[0], palette=colors, showfliers=False)\n",
    "sns.boxplot(x='workload', y='hrv_rmssd', data=df, ax=axes[1], palette=colors, showfliers=False)\n",
    "\n",
    "axes[0].set_ylabel('HR (bpm)')\n",
    "axes[1].set_ylabel('HRV (ms)')\n",
    "\n",
    "axes[0].set_xticklabels(['Decrease', 'Increase'])\n",
    "axes[1].set_xticklabels(['Decrease', 'Increase'])\n",
    "\n",
    "axes[0].set_xlabel('')\n",
    "axes[1].set_xlabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures/fig1ecg.pdf')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "chi23lbw",
   "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": "3.9.16"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "82ae58ca00847926547aabc807806e6cb01e8dbaa50ce6f12d50c344679ba919"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}