399 lines (398 with data), 74.9 kB

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2: Clinical Application\n",
    "\n",
    "### Contents\n",
    "Fill out this notebook as part 2 of your final project submission.\n",
    "\n",
    "**You will have to complete the Code (Load Metadata & Compute Resting Heart Rate) and Project Write-up sections.**  \n",
    "\n",
    "- [Code](#Code) is where you will implement some parts of the **pulse rate algorithm** you created and tested in Part 1 and already includes the starter code.\n",
    "  - [Imports](#Imports) - These are the imports needed for Part 2 of the final project. \n",
    "    - [glob](https://docs.python.org/3/library/glob.html)\n",
    "    - [os](https://docs.python.org/3/library/os.html)\n",
    "    - [numpy](https://numpy.org/)\n",
    "    - [pandas](https://pandas.pydata.org/)\n",
    "  - [Load the Dataset](#Load-the-dataset)  \n",
    "  - [Load Metadata](#Load-Metadata)\n",
    "  - [Compute Resting Heart Rate](#Compute-Resting-Heart-Rate)\n",
    "  - [Plot Resting Heart Rate vs. Age Group](#Plot-Resting-Heart-Rate-vs.-Age-Group)\n",
    "- [Project Write-up](#Project-Write-Up) to describe the clinical significance you observe from the **pulse rate algorithm** applied to this dataset, what ways/information that could improve your results, and if we validated a trend known in the science community. \n",
    "\n",
    "### Dataset (CAST)\n",
    "\n",
    "The data from this project comes from the [Cardiac Arrythmia Suppression Trial (CAST)](https://physionet.org/content/crisdb/1.0.0/), which was sponsored by the National Heart, Lung, and Blood Institute (NHLBI). CAST collected 24 hours of heart rate data from ECGs from people who have had a myocardial infarction (MI) within the past two years.[1] This data has been smoothed and resampled to more closely resemble PPG-derived pulse rate data from a wrist wearable.[2]\n",
    "\n",
    "1. **CAST RR Interval Sub-Study Database Citation** - Stein PK, Domitrovich PP, Kleiger RE, Schechtman KB, Rottman JN. Clinical and demographic determinants of heart rate variability in patients post myocardial infarction: insights from the Cardiac Arrhythmia Suppression Trial (CAST). Clin Cardiol 23(3):187-94; 2000 (Mar)\n",
    "2. **Physionet Citation** - Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals (2003). Circulation. 101(23):e215-e220.\n",
    "\n",
    "-----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code\n",
    "#### Imports\n",
    "\n",
    "When you implement the functions, you'll only need to you use the packages you've used in the classroom, like [Pandas](https://pandas.pydata.org/) and [Numpy](http://www.numpy.org/). These packages are imported for you here. We recommend you don't import other packages outside of the [Standard Library](https://docs.python.org/3/library/) , otherwise the grader might not be able to run your code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import glob as glob\n",
    "import os\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Load the dataset\n",
    "\n",
    "The dataset is stored as [.npz](https://docs.scipy.org/doc/numpy/reference/generated/numpy.savez.html) files. Each file contains roughly 24 hours of heart rate data in the 'hr' array sampled at 1Hz. The subject ID is the name of the file. You will use these files to compute resting heart rate.\n",
    "\n",
    "Demographics metadata is stored in a file called 'metadata.csv'. This CSV has three columns, one for subject ID, age group, and sex. You will use this file to make the association between resting heart rate and age group for each gender.\n",
    "\n",
    "Find the dataset in `../datasets/crisdb/`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "hr_filenames = glob.glob('/data/crisdb/*.npz')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Load Metadata\n",
    "Load the metadata file into a datastructure that allows for easy lookups from subject ID to age group and sex."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "metadata_filename = '/data/crisdb/metadata.csv' #path\n",
    "data = pd.read_csv ('/data/crisdb/metadata.csv')\n",
    "\n",
    "metadata = pd.DataFrame(data) #load metadata into variable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Compute Resting Heart Rate\n",
    "For each subject we want to compute the resting heart rate while keeping track of which age group this subject belongs to. An easy, robust way to compute the resting heart rate is to use the lowest 5th percentile value in the heart rate timeseries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def AgeAndRHR(metadata, filename):\n",
    "\n",
    "    # Load the heart rate timeseries\n",
    "    hr_data = np.load(filename)['hr']\n",
    "    \n",
    "    # Compute the resting heart rate from the timeseries by finding the lowest 5th percentile value in hr_data\n",
    "    rhr = np.percentile(hr_data, 5)\n",
    "\n",
    "    # Find the subject ID from the filename.\n",
    "    subject = os.path.splitext(os.path.basename(filename))[0]\n",
    "\n",
    "    # Find the age group for this subject in metadata.\n",
    "    age_group = metadata.loc[metadata['subject'] == subject,'age'].item()\n",
    "    \n",
    "    # Find the sex for this subject in metadata.\n",
    "    sex = metadata.loc[metadata['subject'] == subject,'sex'].item()\n",
    "\n",
    "    return age_group, sex, rhr\n",
    "\n",
    "df = pd.DataFrame(data=[AgeAndRHR(metadata, filename) for filename in hr_filenames],\n",
    "                  columns=['age_group', 'sex', 'rhr'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plot Resting Heart Rate vs. Age Group\n",
    "We'll use [seaborn](https://seaborn.pydata.org/) to plot the relationship. Seaborn is a thin wrapper around matplotlib, which we've used extensively in this class, that enables higher-level statistical plots.\n",
    "\n",
    "We will use [lineplot](https://seaborn.pydata.org/generated/seaborn.lineplot.html#seaborn.lineplot) to plot the mean of the resting heart rates for each age group along with the 95% confidence interval around the mean. Learn more about making plots that show uncertainty [here](https://seaborn.pydata.org/tutorial/relational.html#aggregation-and-representing-uncertainty)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting package metadata: done\n",
      "Solving environment: done\n",
      "\n",
      "\n",
      "==> WARNING: A newer version of conda exists. <==\n",
      "  current version: 4.6.14\n",
      "  latest version: 4.8.3\n",
      "\n",
      "Please update conda by running\n",
      "\n",
      "    $ conda update -n base conda\n",
      "\n",
      "\n",
      "\n",
      "## Package Plan ##\n",
      "\n",
      "  environment location: /opt/conda\n",
      "\n",
      "  added / updated specs:\n",
      "    - seaborn==0.10.0\n",
      "\n",
      "\n",
      "The following packages will be downloaded:\n",
      "\n",
      "    package                    |            build\n",
      "    ---------------------------|-----------------\n",
      "    ca-certificates-2020.4.5.1 |       hecc5488_0         146 KB  conda-forge\n",
      "    certifi-2020.4.5.1         |   py36h9f0ad1d_0         151 KB  conda-forge\n",
      "    kiwisolver-1.2.0           |   py36hdb11119_0          87 KB  conda-forge\n",
      "    matplotlib-2.2.2           |   py36h0e671d2_0         6.6 MB  defaults\n",
      "    numba-0.35.0               |     np113py36_10         2.6 MB  defaults\n",
      "    numpy-1.13.3               |py36_blas_openblas_200         8.7 MB  conda-forge\n",
      "    python_abi-3.6             |          1_cp36m           4 KB  conda-forge\n",
      "    seaborn-0.10.0             |             py_0         158 KB  conda-forge\n",
      "    ------------------------------------------------------------\n",
      "                                           Total:        18.5 MB\n",
      "\n",
      "The following NEW packages will be INSTALLED:\n",
      "\n",
      "  kiwisolver         conda-forge/linux-64::kiwisolver-1.2.0-py36hdb11119_0\n",
      "  python_abi         conda-forge/linux-64::python_abi-3.6-1_cp36m\n",
      "\n",
      "The following packages will be UPDATED:\n",
      "\n",
      "  ca-certificates                     2019.11.28-hecc5488_0 --> 2020.4.5.1-hecc5488_0\n",
      "  certifi                                 2019.11.28-py36_0 --> 2020.4.5.1-py36h9f0ad1d_0\n",
      "  matplotlib                           2.1.0-py36hba5de38_0 --> 2.2.2-py36h0e671d2_0\n",
      "  numba                 pkgs/free::numba-0.35.0-np112py36_0 --> pkgs/main::numba-0.35.0-np113py36_10\n",
      "  numpy                       1.12.1-py36_blas_openblas_200 --> 1.13.3-py36_blas_openblas_200\n",
      "  seaborn            conda-forge/linux-64::seaborn-0.8.1-p~ --> conda-forge/noarch::seaborn-0.10.0-py_0\n",
      "\n",
      "\n",
      "\n",
      "Downloading and Extracting Packages\n",
      "python_abi-3.6       | 4 KB      | ##################################### | 100% \n",
      "matplotlib-2.2.2     | 6.6 MB    | ##################################### | 100% \n",
      "ca-certificates-2020 | 146 KB    | ##################################### | 100% \n",
      "numba-0.35.0         | 2.6 MB    | ##################################### | 100% \n",
      "seaborn-0.10.0       | 158 KB    | ##################################### | 100% \n",
      "numpy-1.13.3         | 8.7 MB    | ##################################### | 100% \n",
      "certifi-2020.4.5.1   | 151 KB    | ##################################### | 100% \n",
      "kiwisolver-1.2.0     | 87 KB     | ##################################### | 100% \n",
      "Preparing transaction: done\n",
      "Verifying transaction: done\n",
      "Executing transaction: done\n"
     ]
    }
   ],
   "source": [
    "!conda install seaborn==0.10.0 -y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting package metadata: done\n",
      "Solving environment: done\n",
      "\n",
      "\n",
      "==> WARNING: A newer version of conda exists. <==\n",
      "  current version: 4.6.14\n",
      "  latest version: 4.8.3\n",
      "\n",
      "Please update conda by running\n",
      "\n",
      "    $ conda update -n base conda\n",
      "\n",
      "\n",
      "\n",
      "## Package Plan ##\n",
      "\n",
      "  environment location: /opt/conda\n",
      "\n",
      "  added / updated specs:\n",
      "    - seaborn=0.9.0\n",
      "\n",
      "\n",
      "The following packages will be downloaded:\n",
      "\n",
      "    package                    |            build\n",
      "    ---------------------------|-----------------\n",
      "    ca-certificates-2020.1.1   |                0         132 KB  anaconda\n",
      "    certifi-2020.4.5.1         |           py36_0         159 KB  anaconda\n",
      "    openssl-1.0.2u             |       h7b6447c_0         3.1 MB  anaconda\n",
      "    seaborn-0.9.0              |     pyh91ea838_1         164 KB  anaconda\n",
      "    ------------------------------------------------------------\n",
      "                                           Total:         3.6 MB\n",
      "\n",
      "The following packages will be UPDATED:\n",
      "\n",
      "  ca-certificates    conda-forge::ca-certificates-2019.11.~ --> anaconda::ca-certificates-2020.1.1-0\n",
      "  certifi            conda-forge::certifi-2019.11.28-py36_0 --> anaconda::certifi-2020.4.5.1-py36_0\n",
      "  seaborn            conda-forge/linux-64::seaborn-0.8.1-p~ --> anaconda/noarch::seaborn-0.9.0-pyh91ea838_1\n",
      "\n",
      "The following packages will be SUPERSEDED by a higher-priority channel:\n",
      "\n",
      "  openssl            conda-forge::openssl-1.0.2u-h516909a_0 --> anaconda::openssl-1.0.2u-h7b6447c_0\n",
      "\n",
      "The following packages will be DOWNGRADED:\n",
      "\n",
      "  scipy                                1.2.1-py36h09a28d5_1 --> 0.19.1-py36_blas_openblas_202\n",
      "\n",
      "\n",
      "\n",
      "Downloading and Extracting Packages\n",
      "openssl-1.0.2u       | 3.1 MB    | ##################################### | 100% \n",
      "certifi-2020.4.5.1   | 159 KB    | ##################################### | 100% \n",
      "seaborn-0.9.0        | 164 KB    | ##################################### | 100% \n",
      "ca-certificates-2020 | 132 KB    | ##################################### | 100% \n",
      "Preparing transaction: done\n",
      "Verifying transaction: done\n",
      "Executing transaction: done\n"
     ]
    }
   ],
   "source": [
    "!conda install -y -c anaconda seaborn=0.9.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1e3d9eb438>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "#lineplot\n",
    "labels = sorted(np.unique(df.age_group))\n",
    "df['xaxis'] = df.age_group.map(lambda x: labels.index(x)).astype('float')\n",
    "plt.figure(figsize=(12, 8))\n",
    "sns.lineplot(x='xaxis', y='rhr', hue='sex', data=df)\n",
    "_ = plt.xticks(np.arange(len(labels)), labels)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clinical Conclusion\n",
    "Answer the following prompts to draw a conclusion about the data.\n",
    "> 1. For women, we see .... \n",
    "> 2. For men, we see ... \n",
    "> 3. In comparison to men, women's heart rate is .... \n",
    "> 4. What are some possible reasons for what we see in our data?\n",
    "> 5. What else can we do or go and find to figure out what is really happening? How would that improve the results?\n",
    "> 6. Did we validate the trend that average resting heart rate increases up until middle age and then decreases into old age? How?\n",
    "\n",
    "Your write-up will go here..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **1**. For women, we see that the heart rate starts to increase from age group (35-39) to age group (40-44) and then decreases from age group (40-44) to age group (70-74). After that, it rises again slowly.\n",
    "\n",
    "> **2**. For men, we see a steady heart rate for 35 to 75. The decrease compared to women is smaller. There is also an increasing trend for age groups(40-49) and (55-69)\n",
    "\n",
    "> **3**. In comparison to men, women's heart rate has higher variance. In addition, it's higher while resting.\n",
    "\n",
    "> **4**. The possible reasons are: the data contains people who had myocardial infection (MI) and women sample is liwer than men (277 by 1260). In addition, the sample might be from an specific demographic characteristic.\n",
    "\n",
    "> **5**. We should repeat the experiment with greater dataset (and less biased). \n",
    "\n",
    "> **6**. This happens in women's sample. The average resting rate increases up until midle age and then gradually decreases. For men the decrease was smaller. However, we need to use more data to validate this hypothesis. \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}