taylor_2000_accelerations_decelerations.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Accelerations and decelerations from Taylor et al. 2000\n",
    "\n",
    "## Summary\n",
    "\n",
    "This notebook implements the method for acceleration and deceleration detection from Taylor et al. 2000, based on the MATLAB implemention in the FHRMA package.\n",
    "\n",
    "[Taylor et al. 2000](https://doi.org/10.1111/j.1471-0528.2000.tb11112.x) - The development and validation of an algorithm for real-time computerised fetal heart rate monitoring in labour\n",
    "\n",
    "## Paper description of method\n",
    "\n",
    "**Cleaning signals:** \"Both the fetal heart rate and uterine pressure signals contained artefact... Artefact must be removed prior to determining the timing of the decelerations; this is achieved by a low pass fourth order Butterworth filter with a cut off frequency of 0.02 Hz\"\n",
    "\n",
    "**Identifying accelerations and decelerations:** \"Accelerations and decelerations were determined and classified according to definitions given by FIGO.\"\n",
    "\n",
    "Not performed in FHRMA but also in paper:\n",
    "* **Classifying decelerations:** \"The classification of decelerations was based upon the temporal relationship between the minimum of the deceleration and the maximum of the associated uterine contraction. The uterine contractions were timed by means of an external tocograph. A late deceleration occurred where its minimum value was 20–60 s after the peak of a contraction\"\n",
    "* **Calculating FHR variability:** \"The fetal heart rate variability is calculated as twice the standard deviation of the fetal heart rate recorded continuously, after removing the accelerations and decelerations. Reduced variability is defined as one standard deviation < 5 beats per minute.\"\n",
    "\n",
    "As you can see, the description of the method for identifying accelerations and decelerations is not clear, and I have followed the interpretation of Boudet et al. in their MATLAB code."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MATLAB Implementation\n",
    "\n",
    "FHRMA includes results from using the simple acceleration/deceleration algorithm, as well as results from their interpretation of Taylor et al. 2000. The Taylor version uses `aamtaylor.m`, `butterfilt.m`, `startendlist.m`, and `validaccident.m`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set-up\n",
    "\n",
    "Import required packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from scipy.signal import butter, sosfiltfilt\n",
    "import sys"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import functions defined elsewhere"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sys.path.append('/functions')\n",
    "from functions.load_data import load_data\n",
    "from functions.simple_acc_dec_detection import create_accident_df\n",
    "from functions.simple_acc_dec_validation import validation_against_fhrma"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Python implementation of the paper's method\n",
    "\n",
    "Import datasets required for running the function and validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_fhr, fhrma_base, fhrma_acc, fhrma_dec = load_data(fhrma_path = './data/T_orig.mat')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Result from FHRMA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([], shape=(0, 2), dtype=uint8)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fhrma_acc['train28']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step-by-step implementation in Python**\n",
    "\n",
    "Apply Butterworth filter to the FHR data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "fhr = raw_fhr['train28']\n",
    "baseline = fhrma_base['train28']\n",
    "\n",
    "# Generate butterworth filter\n",
    "sos = butter(\n",
    "    N=4, # Filter's order\n",
    "    Wn=0.002, # Critical frequency\n",
    "    btype='low', # Kind of filter\n",
    "    fs = 4, # Sampling frequency\n",
    "    output='sos' # Output type\n",
    ")\n",
    "\n",
    "# Use filtfilt() (instead of sosfilt) as it applies the filter\n",
    "# twice, forwards and backwards, so you get zero-phase filtering\n",
    "fuzzy_line = sosfiltfilt(sos, fhr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compared to MATLAB, the fuzzy line is very flat, so not anticipating will match up with FHRMA. In order to be more consistent, would need to produce a butterworth filter manually that matches theirs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f97d98a1f00>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show what this has done\n",
    "plt.plot(fuzzy_line)\n",
    "plt.plot(baseline)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create list of periods where there is a 10bpm difference - between baseline and the fuzzyline - these are the acceleration candidates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    17595\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binsig = (fuzzy_line-baseline > 10)\n",
    "pd.Series(binsig).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Not continued as not going to match up without matching butterworth."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Python implementation of the simple acc/dec detection method\n",
    "\n",
    "Import datasets required for running the function and validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_fhr, fhrma_base, fhrma_acc, fhrma_dec = load_data(fhrma_path = './data/T_std.mat')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Accelerations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>record</th>\n",
       "      <th>fhrma</th>\n",
       "      <th>python</th>\n",
       "      <th>match</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>train28</td>\n",
       "      <td>[[61.8625, 62.31666666666667]]</td>\n",
       "      <td>[[61.8625, 62.31666666666667]]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test80</td>\n",
       "      <td>[[21.7875, 22.241666666666667], [25.7791666666...</td>\n",
       "      <td>[[21.7875, 22.241666666666667], [25.7791666666...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>test68</td>\n",
       "      <td>[[8.545833333333333, 8.9625], [9.125, 9.4875],...</td>\n",
       "      <td>[[8.545833333333333, 8.9625], [9.125, 9.4875],...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>train04</td>\n",
       "      <td>[[0.05, 0.31666666666666665], [25.65, 26.30416...</td>\n",
       "      <td>[[0.05, 0.31666666666666665], [25.65, 26.30416...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>train31</td>\n",
       "      <td>[[16.9125, 17.979166666666668], [19.2, 19.5916...</td>\n",
       "      <td>[[16.9125, 17.979166666666668], [19.2, 19.5916...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    record                                              fhrma  \\\n",
       "0  train28                     [[61.8625, 62.31666666666667]]   \n",
       "1   test80  [[21.7875, 22.241666666666667], [25.7791666666...   \n",
       "2   test68  [[8.545833333333333, 8.9625], [9.125, 9.4875],...   \n",
       "3  train04  [[0.05, 0.31666666666666665], [25.65, 26.30416...   \n",
       "4  train31  [[16.9125, 17.979166666666668], [19.2, 19.5916...   \n",
       "\n",
       "                                              python  match  \n",
       "0                     [[61.8625, 62.31666666666667]]   True  \n",
       "1  [[21.7875, 22.241666666666667], [25.7791666666...   True  \n",
       "2  [[8.545833333333333, 8.9625], [9.125, 9.4875],...   True  \n",
       "3  [[0.05, 0.31666666666666665], [25.65, 26.30416...   True  \n",
       "4  [[16.9125, 17.979166666666668], [19.2, 19.5916...   True  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count_diff\n",
      "0    156\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total_diff\n",
      "0.0    156\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc_result = create_accident_df(\n",
    "    raw_fhr=raw_fhr,\n",
    "    fhrma_base=fhrma_base,\n",
    "    fhrma_dict=fhrma_acc,\n",
    "    type='acc')\n",
    "display(acc_result.head())\n",
    "\n",
    "validation_against_fhrma(acc_result, 'accelerations')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decelerations"
   ]
  },
  {
   "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>record</th>\n",
       "      <th>fhrma</th>\n",
       "      <th>python</th>\n",
       "      <th>match</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>train28</td>\n",
       "      <td>[[9.2625, 9.941666666666666], [46.0, 46.254166...</td>\n",
       "      <td>[[9.2625, 9.941666666666666], [46.0, 46.254166...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test80</td>\n",
       "      <td>[[67.40416666666667, 67.69583333333334], [68.3...</td>\n",
       "      <td>[[67.40416666666667, 67.69583333333334], [68.3...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>test68</td>\n",
       "      <td>[[26.033333333333335, 26.625], [34.15416666666...</td>\n",
       "      <td>[[26.033333333333335, 26.625], [34.15416666666...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>train04</td>\n",
       "      <td>[[3.9125, 4.179166666666666], [23.3875, 24.141...</td>\n",
       "      <td>[[3.9125, 4.179166666666666], [23.3875, 24.141...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>train31</td>\n",
       "      <td>[[0.15, 0.6291666666666667], [3.7625, 4.379166...</td>\n",
       "      <td>[[0.15, 0.6291666666666667], [3.7625, 4.379166...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    record                                              fhrma  \\\n",
       "0  train28  [[9.2625, 9.941666666666666], [46.0, 46.254166...   \n",
       "1   test80  [[67.40416666666667, 67.69583333333334], [68.3...   \n",
       "2   test68  [[26.033333333333335, 26.625], [34.15416666666...   \n",
       "3  train04  [[3.9125, 4.179166666666666], [23.3875, 24.141...   \n",
       "4  train31  [[0.15, 0.6291666666666667], [3.7625, 4.379166...   \n",
       "\n",
       "                                              python  match  \n",
       "0  [[9.2625, 9.941666666666666], [46.0, 46.254166...   True  \n",
       "1  [[67.40416666666667, 67.69583333333334], [68.3...   True  \n",
       "2  [[26.033333333333335, 26.625], [34.15416666666...   True  \n",
       "3  [[3.9125, 4.179166666666666], [23.3875, 24.141...   True  \n",
       "4  [[0.15, 0.6291666666666667], [3.7625, 4.379166...   True  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count_diff\n",
      "0    156\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total_diff\n",
      "0.0    156\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dec_result = create_accident_df(\n",
    "    raw_fhr=raw_fhr,\n",
    "    fhrma_base=fhrma_base,\n",
    "    fhrma_dict=fhrma_dec,\n",
    "    type='dec')\n",
    "display(dec_result.head())\n",
    "\n",
    "validation_against_fhrma(dec_result, 'decelerations')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ctg",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}