Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
COVID19.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "COVID19.ipynb",
"provenance": [],
"collapsed_sections": [],
"authorship_tag": "ABX9TyMyqhIP/9NjgjYm0GK56SZT",
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/irishryoon/fe2c769ccca1028dbda2e4a5f060fb7e/covid19.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "IgTpCRCvVWK-",
"colab_type": "text"
},
"source": [
"# When will COVID-19 take over US hospitals?\n",
"* This notebook shows how one can fit the SIR model to predict the dates during which US hospitals will experience bed shortage. \n",
"* An app has been developed based on the work presented in this notebook. Check out the app: https://covid19-hospital.herokuapp.com/ \n",
"* This notebook is organized according to the following sections: \n",
" 0. Define Functions\n",
" 1. Get data\n",
" 2. Model performance in Hubei, china \n",
" 3. Analysis by state\n",
" 4. Analysis by county (for State of Washington)\n",
"\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "jUEhbFfeLVKV",
"colab_type": "code",
"outputId": "8de6a798-53f4-4f1b-a638-c2ebfb42f9f1",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 154
}
},
"source": [
"# Install symfit: https://symfit.readthedocs.io/en/latest/fitting_types.html#ode-fitting\n",
"!pip install symfit\n",
"# for selenium\n",
"!apt update\n",
"!apt install chromium-chromedriver\n",
"!pip install selenium"
],
"execution_count": 1,
"outputs": [
{
"output_type": "stream",
"text": [
"Requirement already satisfied: symfit in /usr/local/lib/python3.6/dist-packages (0.5.2)\n",
"Requirement already satisfied: sympy>=1.2 in /usr/local/lib/python3.6/dist-packages (from symfit) (1.5.1)\n",
"Requirement already satisfied: numpy>=1.12 in /usr/local/lib/python3.6/dist-packages (from symfit) (1.18.2)\n",
"Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from symfit) (1.12.0)\n",
"Requirement already satisfied: toposort in /usr/local/lib/python3.6/dist-packages (from symfit) (1.5)\n",
"Requirement already satisfied: scipy<1.2,>=1.0 in /usr/local/lib/python3.6/dist-packages (from symfit) (1.1.0)\n",
"Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.6/dist-packages (from sympy>=1.2->symfit) (1.1.0)\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "KLi3p-zvPCgG",
"colab_type": "code",
"colab": {}
},
"source": [
"from selenium import webdriver\n",
"import regex as re\n",
"import glob\n",
"import datetime\n",
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"from matplotlib import style\n",
"plt.style.use('ggplot')\n",
"\n",
"import pickle \n",
"from collections import OrderedDict\n",
"from bs4 import BeautifulSoup\n",
"from urllib.error import HTTPError\n",
"from urllib import request, response, error, parse\n",
"from urllib.request import urlopen\n",
"\n",
"from symfit import Parameter, variables, Fit, D, ODEModel"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "Q3CmKAlpGDBk",
"colab_type": "text"
},
"source": [
"# 0. Define functions "
]
},
{
"cell_type": "code",
"metadata": {
"id": "cJGw2GnBHAMK",
"colab_type": "code",
"colab": {}
},
"source": [
"abbreviation = {\n",
" 'AK': 'Alaska',\n",
" 'AL': 'Alabama',\n",
" 'AR': 'Arkansas',\n",
" 'AS': 'American Samoa',\n",
" 'AZ': 'Arizona',\n",
" 'CA': 'California',\n",
" 'CO': 'Colorado',\n",
" 'CT': 'Connecticut',\n",
" 'DC': 'District of Columbia',\n",
" 'DE': 'Delaware',\n",
" 'FL': 'Florida',\n",
" 'GA': 'Georgia',\n",
" 'GU': 'Guam',\n",
" 'HI': 'Hawaii',\n",
" 'IA': 'Iowa',\n",
" 'ID': 'Idaho',\n",
" 'IL': 'Illinois',\n",
" 'IN': 'Indiana',\n",
" 'KS': 'Kansas',\n",
" 'KY': 'Kentucky',\n",
" 'LA': 'Louisiana',\n",
" 'MA': 'Massachusetts',\n",
" 'MD': 'Maryland',\n",
" 'ME': 'Maine',\n",
" 'MI': 'Michigan',\n",
" 'MN': 'Minnesota',\n",
" 'MO': 'Missouri',\n",
" 'MP': 'Northern Mariana Islands',\n",
" 'MS': 'Mississippi',\n",
" 'MT': 'Montana',\n",
" 'NA': 'National',\n",
" 'NC': 'North Carolina',\n",
" 'ND': 'North Dakota',\n",
" 'NE': 'Nebraska',\n",
" 'NH': 'New Hampshire',\n",
" 'NJ': 'New Jersey',\n",
" 'NM': 'New Mexico',\n",
" 'NV': 'Nevada',\n",
" 'NY': 'New York',\n",
" 'OH': 'Ohio',\n",
" 'OK': 'Oklahoma',\n",
" 'OR': 'Oregon',\n",
" 'PA': 'Pennsylvania',\n",
" 'PR': 'Puerto Rico',\n",
" 'RI': 'Rhode Island',\n",
" 'SC': 'South Carolina',\n",
" 'SD': 'South Dakota',\n",
" 'TN': 'Tennessee',\n",
" 'TX': 'Texas',\n",
" 'UT': 'Utah',\n",
" 'VA': 'Virginia',\n",
" 'VI': 'Virgin Islands',\n",
" 'VT': 'Vermont',\n",
" 'WA': 'Washington',\n",
" 'WI': 'Wisconsin',\n",
" 'WV': 'West Virginia',\n",
" 'WY': 'Wyoming'}"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "AawrdAle9zFj",
"colab_type": "code",
"colab": {}
},
"source": [
"def get_state_df(data, state, state_name):\n",
" # create dataframe for selected state\n",
" \"\"\"\n",
" --- input ---\n",
" data: (dic) \n",
" state: (str) state abbreviation\n",
" state_name: (str) state name given by abbreviation[state]\n",
"\n",
" --- output ---\n",
" state_df: (DataFrame)\n",
" \"\"\"\n",
"\n",
" # NOTE: in \"data\"\n",
" # On some days, the \"Province/State\" is reported as \"County, State\"\n",
" # On some days, the Province/State\" is reported as \"State\"\n",
"\n",
" keys = list(data.keys())\n",
" keys.sort()\n",
" last = keys[-1]\n",
" columns = data[last].columns\n",
"\n",
" state_data = pd.DataFrame(columns = columns) \n",
"\n",
" # get state data from different days\n",
" for date in keys:\n",
" df = data[date]\n",
" day_data = df[(df['Country/Region'] == 'US') & \n",
" ((df['Province/State'] == state_name) | (df['Province/State'].str.contains(state))\n",
" )]\n",
" state_data = state_data.append(day_data, ignore_index = True)\n",
"\n",
" # sort data, from oldest (top) to newest (bottom) \n",
" state_data = state_data.sort_values(by = [\"report date\"])\n",
"\n",
" # Group by: make sure that all reported under \"County, State\" are aggregated to one row with \"State\"\n",
" state_df = pd.DataFrame(columns = state_data.columns)\n",
"\n",
" state_df[\"Confirmed\"] = state_data.groupby([\"report date\"])[\"Confirmed\"].sum().values\n",
" state_df[\"Deaths\"] = state_data.groupby([\"report date\"])[\"Deaths\"].sum().values\n",
" state_df[\"Recovered\"] = state_data.groupby([\"report date\"])[\"Recovered\"].sum().values\n",
" state_df[\"Province/State\"] = state_name\n",
" state_df[\"Country/Region\"] = \"US\"\n",
" state_df[\"report date\"] = state_data.groupby([\"report date\"])[\"Confirmed\"].sum().keys()\n",
" \n",
" # Simulate the number of recovered people on each day\n",
" # And adjust \"Confirmed\" to report only the number of active cases\n",
" # \"Infected\" is the number of active cases\n",
" n_days = len(state_df[\"Confirmed\"])\n",
" # remove number of deaths from confirmed cases \n",
" infected = [state_df[\"Confirmed\"][0] - state_df[\"Deaths\"][0]]\n",
" recovered_sim = [0]\n",
"\n",
" for i in range(1, n_days):\n",
" R_day = int(sum(infected[:i]) * 0.02)\n",
" recovered_sim.append(R_day)\n",
" infected.append(state_df.at[i,\"Confirmed\"] - R_day)\n",
"\n",
" state_df[\"Infected\"] = infected\n",
" state_df[\"death_or_recovered\"] = recovered_sim\n",
"\n",
" return state_df"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "45HXFo64B6s8",
"colab_type": "code",
"colab": {}
},
"source": [
"def check_data_sufficiency(state_df):\n",
" # Check if the selected state has enough data for the model. \n",
" # Check if the state has 5 consecutive days satisfying the following:\n",
" # (a) The number of confirmed cases are all above 5\n",
" # (b) The number of confirmed cases are increasing every day\n",
" \"\"\"\n",
" --- input ---\n",
" state_df: (DataFrame) output of function \"get_state_df\"\n",
"\n",
" --- output ---\n",
" sufficiency = (bool) True if there is enough data\n",
" first_day = (str if sufficiency == True, None if sufficiency == False) \n",
" (str) The earliest day of the possible 5 consecutive days\n",
" last_day = (str if sufficiency == True, None if sufficiency == False) \n",
" (str) The last available day of data \n",
" \"\"\"\n",
"\n",
" infected = state_df[\"Infected\"].values\n",
" dates = state_df[\"report date\"].values\n",
"\n",
" first_day_found = False\n",
" for idx, I in enumerate(infected):\n",
" if I >= 5:\n",
" # check if the number of infected people increases for 5 consecutive days \n",
" next_days = infected[idx:idx+6]\n",
"\n",
" # check if the numbers are increasing:\n",
" if np.all(np.diff(next_days )) > 0:\n",
" first_day_found = True\n",
" break\n",
" \n",
" if first_day_found == False:\n",
" print(\"Insufficient data for state \", state)\n",
"\n",
" sufficient = first_day_found\n",
" first_day = dates[idx]\n",
" last_day = dates[-1]\n",
"\n",
" return sufficient, first_day, last_day "
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "42vGB6y_8qv1",
"colab_type": "code",
"colab": {}
},
"source": [
"def fit_model(state, state_df, first_day, last_day, n_susceptible, n_days):\n",
" # Take state data to fit the SIR model\n",
" # Use the fitted model to make predictions for the next \"n_days\"\n",
"\n",
" \"\"\"\n",
" --- input ---\n",
" state: (str) selected state. \n",
" state_df: (dataframe) of state data. output from function \"get_state_df\"\n",
" first_day: (str) first day of data to use for model fitting. output of function \"check_data_sufficiency\"\n",
" last_day: (str) last day of data to use for model fitting. output of function \"check_data_sufficiency\"\n",
" NOTE: only data from \"first_day\" to \"last_day\" will be used in the parameter tuning process\n",
" n_susceptible: (int) number of susceptible population \n",
" n_days: (int) number of days to make predictions. \n",
" --- output ---\n",
" prediction: (dict) of prediction info\n",
" \"\"\"\n",
" \n",
" ### get data in correct format\n",
" data_duration = (datetime.datetime.strptime(last_day, \"%m-%d-%Y\") - datetime.datetime.strptime(first_day, \"%m-%d-%Y\")).days + 1 \n",
" data_time = np.array(range(0,data_duration)).astype(int) # time 0 corresponds to the first day appearing in dataframe\n",
" first_day_idx = state_df[state_df[\"report date\"] == first_day].index[0]\n",
" last_day_idx = state_df[state_df[\"report date\"] == last_day].index[0]\n",
"\n",
" data_I= np.array(state_df.loc[first_day_idx: last_day_idx, \"Infected\"])\n",
" data_R = np.array(state_df.loc[first_day_idx:last_day_idx, \"death_or_recovered\"])\n",
" data_S = [n_susceptible - x - data_R[idx] for idx, x in enumerate(data_I)]\n",
"\n",
" ### define variables \n",
" S, I, R, t = variables('S, I, R, t')\n",
"\n",
" ### find sensible initial parameter values\n",
" dIdt = np.diff(data_I)\n",
" dRdt = np.diff(data_R)\n",
" gamma_0 = 0.02\n",
" beta_0 = np.mean([(x + gamma_0 * data_I[idx] ) * n_susceptible / (data_S[idx] * data_I[idx]) for idx, x in enumerate(dIdt)])\n",
" \n",
" ### define parameters\n",
" beta = Parameter('beta', beta_0)\n",
" gamma = Parameter(\"gamma\", gamma_0)\n",
"\n",
" ### define ODE equations \n",
" model_dict = {\n",
" D(S, t): - beta * S * I / n_susceptible,\n",
" D(I, t): beta * S* I / n_susceptible - gamma * I,\n",
" D(R, t): gamma * I}\n",
"\n",
" ### set initial values\n",
" I0 = data_I[0]\n",
" S0 = n_susceptible - I0\n",
" R0 = data_R[0]\n",
"\n",
" ### define the model\n",
" model = ODEModel(model_dict, initial = { t : 0, S : S0, I : I0, R : R0 })\n",
"\n",
" ### fit model parameters\n",
" fit = Fit(model, t = data_time, I = data_I, S = None, R = data_R )\n",
" fit_result = fit.execute()\n",
"\n",
" ### make sure the parameters make sense\n",
" # Ocassionaly (when there isn't enough data), parameters may be non-positive.\n",
" # If this is the case, use the naive guess from data\n",
" params = fit_result.params\n",
" if params[\"beta\"] <= 0 or params[\"gamma\"] <=0 :\n",
" params = OrderedDict()\n",
" params[\"beta\"] = beta_0\n",
" params[\"gamma\"] = gamma_0\n",
" if params[\"gamma\"] >= 0.1:\n",
" params[\"beta\"] = beta_0\n",
" params[\"gamma\"] = gamma_0\n",
"\n",
" ### get predictions\n",
" tvec = np.linspace(0, n_days-1, n_days)\n",
" outcome = model(t=tvec, **params)\n",
" I_pred = outcome.I\n",
" S_pred = outcome.S\n",
" R_pred = outcome.R \n",
"\n",
" ### save predictions\n",
" prediction = {}\n",
" prediction[\"state\"] = state\n",
" prediction[\"first_day\"] = first_day\n",
" prediction[\"last_day\"] = last_day\n",
" prediction[\"num_days\"] = n_days\n",
" prediction[\"susceptible\"] = n_susceptible\n",
" prediction[\"infected_pred\"] = I_pred\n",
" prediction[\"susceptible_pred\"] = S_pred\n",
" prediction[\"recovered_pred\"] = R_pred\n",
" prediction[\"parameters\"] = params\n",
"\n",
" return prediction"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "RUXhE5LUqScN",
"colab_type": "code",
"colab": {}
},
"source": [
"def predict_by_age(age_statistic, age_prediction, column_prefix):\n",
"\n",
" # predict the number of hospitalization and ICU admission in each age group. \n",
"\n",
" \"\"\"\n",
" --- input ---\n",
" age_statistic: (array) of statistic on either\n",
" (1) hospital admission among infected people by age group, or\n",
" (2) ICU admission among infected people by age group\n",
" age_prediction: (DataFrame) prediction of number of infected people by age group\n",
" Must have the columns [\"0-19\",\"20-44\",\"45-54\",\"55-64\",\"65-74\",\"75-84\",\"85+\"]\n",
" column_prefix: (str) prefix to column\n",
"\n",
" --- output ---\n",
" df :(DataFrame) of prediction by age group and prediction total\n",
" \"\"\"\n",
" \n",
" age_cols = [\"0-19\",\"20-44\",\"45-54\",\"55-64\",\"65-74\",\"75-84\",\"85+\"] \n",
" n_rows = age_prediction.shape[0]\n",
"\n",
" # Make prediction in each age group\n",
" m = np.repeat(age_statistic, [n_rows], axis = 0)\n",
" pred = np.multiply(age_prediction.values, m) \n",
" \n",
" columns = [column_prefix + x for x in age_cols]\n",
" df = pd.DataFrame(pred, columns = columns)\n",
" \n",
" # Make total prediction\n",
" df[column_prefix + \"_total\"] = df.sum(axis = 1)\n",
"\n",
" return df\n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "xKnfgza-DQZ4",
"colab_type": "code",
"colab": {}
},
"source": [
"def predict_hospitalization_ICU(region_dem, P):\n",
"\n",
" \"\"\"\n",
" --- input ---\n",
" region_dem: (DataFrame) of demographic information of particular state or county.\n",
" Must have columns [\"per0-19\",\"per20-44\", \"per45-54\", \"per55-64\",\"per65-74\", \"per75-84\", \"per85plus\"]\n",
" that indicate the percentage of particular age group \n",
" P: (dict) prediction outcome of function \"train_model\".\n",
" Must have the key \"infected_pred\" and \"num_days\"\n",
"\n",
" --- output ---\n",
" P_age: (dict) prediction of number of hospital and ICU admissions \n",
"\n",
" \"\"\"\n",
"\n",
" ### Statistic for hospital admission by age group ###\n",
" # 0. data from CDC: https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e2.htm?s_cid=mm6912e2_w#T1_down\n",
" age_cols = [\"0-19\",\"20-44\",\"45-54\",\"55-64\",\"65-74\",\"75-84\",\"85+\"] \n",
"\n",
" hospitalization_max = np.array([0.025, 0.208, 0.283, 0.301, 0.435, 0.587, 0.703]).reshape(1,-1)\n",
" hospitalization_min = np.array([0.016, 0.143, 0.212, 0.205, 0.286, 0.305, 0.313]).reshape(1,-1)\n",
" ICU_max = np.array([0. , 0.042, 0.104, 0.112, 0.188, 0.31 , 0.29 ]).reshape(1,-1)\n",
" ICU_min = np.array([0. , 0.02 , 0.054, 0.047, 0.081, 0.105, 0.063]).reshape(1,-1)\n",
"\n",
" first_dt = datetime.datetime.strptime(P[\"first_day\"], \"%m-%d-%Y\")\n",
"\n",
" # 1. create dataframe P_age \n",
" P_age = pd.DataFrame(columns = [\"date\", \"total\"] + age_cols)\n",
" P_age[\"total\"] = P[\"infected_pred\"]\n",
" n_rows = P[\"num_days\"]\n",
" P_age[\"date\"] = [(first_dt + datetime.timedelta(days = x)).strftime(\"%m-%d-%Y\") for x in range(0,n_rows)]\n",
"\n",
" # 2. Predict the number of infected people in each age group\n",
" for idx in P_age.index:\n",
" total = P_age.at[idx,\"total\"]\n",
" P_age.loc[idx, age_cols] = (total * region_dem[[\"per0-19\",\"per20-44\", \"per45-54\", \"per55-64\",\"per65-74\", \"per75-84\", \"per85plus\"]]).values[0]\n",
"\n",
" # 3. Predict number of hospital admission using \"hospitalization_min\"\n",
" hos_min_df = predict_by_age(hospitalization_min, P_age[age_cols], \"hos_min\")\n",
" P_age = pd.concat([P_age, hos_min_df], axis = 1)\n",
"\n",
" # 4. Predict number of hospital amidssion using \"hospitalization_max\"\n",
" hos_max_df = predict_by_age(hospitalization_max, P_age[age_cols], \"hos_max\")\n",
" P_age = pd.concat([P_age, hos_max_df], axis = 1)\n",
"\n",
" # 5. Predict number of ICU amidssion using \"ICU_min\"\n",
" ICU_min_df = predict_by_age(ICU_min, P_age[age_cols], \"ICU_min\")\n",
" P_age = pd.concat([P_age, ICU_min_df], axis = 1)\n",
"\n",
" # 6. Predict number of ICU amidssion using \"ICU_max\"\n",
" ICU_max_df = predict_by_age(ICU_max, P_age[age_cols], \"ICU_max\")\n",
" P_age = pd.concat([P_age, ICU_max_df], axis = 1)\n",
"\n",
" return P_age"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "B3kTj10s0PbD",
"colab_type": "code",
"colab": {}
},
"source": [
"def hospital_prediction(hos_prediction, n_beds, region):\n",
"# predict when the hospital will run out of beds and how many extra beds hospitals need\n",
" \"\"\"\n",
" --- input ---\n",
" hos_prediciton: (DataFrame) output of function \"predict_hospitalization_ICU\"\n",
" n_beds: (int) number of available beds\n",
" region: (str) region name. ex) \"WA\" \n",
" \n",
" --- output ---\n",
" None\n",
" \"\"\"\n",
" last_prediction_day = hos_prediction[\"date\"].values[-1]\n",
"\n",
" # Conservative estimate \n",
"\n",
" over = hos_prediction[hos_prediction[\"hos_min_total\"] > n_beds]\n",
"\n",
" min_peak_idx = (hos_prediction[\"hos_min_total\"]).idxmax()\n",
" min_peak = hos_prediction.at[min_peak_idx, \"date\"]\n",
"\n",
" if over.empty == False:\n",
" min_day = over[\"date\"].values[0]\n",
" last_day = over[\"date\"].values[-1]\n",
"\n",
" if last_day < last_prediction_day:\n",
" print(\"According to a conservative estimate, hospitals in \", region, \" will need more beds from \", \n",
" min_day, \" to \", last_day)\n",
" else:\n",
" print(\"According to a conservative estimate, hospitals in \", region, \" will need more beds from \", \n",
" min_day, \" to \", last_day,\" and beyond.\")\n",
" \n",
" min_beds = int(hos_prediction.at[min_peak_idx, \"hos_min_total\"] - n_beds)\n",
" print(\"According to a conservative estimate, hospitals in \", region, \" will experience the highest level of bed shortage on \" \n",
" + min_peak + \", and they will need \" + str(min_beds) + \" extra beds.\")\n",
"\n",
" if over.empty == True:\n",
" min_beds = int(hos_prediction.at[min_peak_idx, \"hos_min_total\"])\n",
" print(\"According to a conservative estimate, hospitals in \", region, \" will see the highest number of COVID-19 patients on \" \n",
" + min_peak + \", and they will need \" +str(min_beds) + \" beds dedicated to COVID-19 patients.\")\n",
"\n",
" # Liberal estimate\n",
" over = hos_prediction[hos_prediction[\"hos_max_total\"] > n_beds]\n",
" max_peak_idx = (hos_prediction[\"hos_max_total\"]).idxmax()\n",
" max_peak = hos_prediction.at[max_peak_idx, \"date\"]\n",
" if over.empty == False:\n",
" max_day = over[\"date\"].values[0]\n",
" last_day = over[\"date\"].values[-1]\n",
" \n",
" if last_day < last_prediction_day:\n",
" print(\"According to a liberal estimate, hospitals in \", region, \" will need more beds from \", \n",
" max_day, \" to \", last_day)\n",
" else:\n",
" print(\"According to a liberal estimate, hospitals in \", region, \" will need more beds from \", \n",
" max_day, \" to \", last_day, \" and beyond.\")\n",
"\n",
" max_beds = int(hos_prediction.at[max_peak_idx, \"hos_max_total\"] - n_beds)\n",
" print(\"According to a liberal estimate, hospitals in \", region, \" will experience the highest level of bed shortage on \" \n",
" + max_peak + \", and they will need \" + str(max_beds) + \" extra beds.\")\n",
"\n",
" if over.empty == True:\n",
" max_beds = int(hos_prediction.at[max_peak_idx, \"hos_max_total\"])\n",
" print(\"According to a liberal estimate, hospitals in \", region, \" will see the highest number of COVID-19 patients on \" \n",
" + max_peak + \", and they will need \" +str(max_beds) + \" beds dedicated to COVID-19 patients.\")"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "uuIigDke-Fj9",
"colab_type": "code",
"colab": {}
},
"source": [
"def plot_pred(ax, P, data_df, plot_title):\n",
" # plot the prediction \n",
" \"\"\"\n",
" --- input ---\n",
" ax: plot object\n",
" P: (dict) output from function \"fit_model\"\n",
" data_df: (DataFrame) of COVID-data for selected region (country / state / county )\n",
" Must have columns \"Infected\", \"Deaths\", \"Recovered\", \"Recovered_sim\", \"report date\"\n",
" There must be one row for each \"report date\" \n",
" plot_title: (str) title of string \n",
"\n",
" --- output ---\n",
" plot \n",
" \"\"\"\n",
"\n",
" # load info from prediction \n",
" first_day = P[\"first_day\"]\n",
" last_day = P[\"last_day\"]\n",
" first_dt = datetime.datetime.strptime(P[\"first_day\"], \"%m-%d-%Y\")\n",
" last_dt = datetime.datetime.strptime(P[\"last_day\"], \"%m-%d-%Y\")\n",
" n_days = P[\"num_days\"] \n",
" tvec = np.linspace(0, n_days-1, n_days)\n",
" data_duration = (last_dt - first_dt).days + 1\n",
" \n",
" # plot predictions\n",
" ax.plot(tvec, P[\"infected_pred\"], label='prediction: infected', c = '#F95858' ) \n",
" ax.plot(tvec, P[\"recovered_pred\"], label='prediction: deceased or recovered ', c = '#6288E2') \n",
"\n",
" # plot data\n",
" data_df = data_df[(data_df[\"report date\"] >= first_day) & (data_df[\"report date\"] <= last_day)]\n",
" ax.scatter(range(0, data_duration), \n",
" data_df[\"Infected\"].values, \n",
" marker = 'o', \n",
" label = 'data: infected', \n",
" c = '#920808')\n",
"\n",
" ax.scatter(range(0, data_duration), \n",
" data_df[\"death_or_recovered\"].values, \n",
" marker = 'o', \n",
" label = 'data: deceased or recovered', \n",
" c = '#244798')\n",
"\n",
" # set labels\n",
" time_label = [(first_dt + datetime.timedelta(days = i)).date().strftime('%m-%d') for i in range(0,n_days,20)]\n",
" ax.set_xticks(list(range(0,n_days,20))) \n",
" ax.set_xticklabels(time_label)\n",
" ax.set_xlabel(\"date\")\n",
" ax.set_ylabel(\"number of people\")\n",
" ax.set_title(plot_title)\n",
"\n",
" return ax"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "mT7Lt7lxGVuv",
"colab_type": "code",
"colab": {}
},
"source": [
"def plot_hospital_pred(ax, P, pred_infected, hospital_capacity, plot_title):\n",
" # plot predictions of hospitalization and ICU admission\n",
" \"\"\"\n",
" --- input --- \n",
" ax: plot object\n",
" P: (dict) output of function \"fit_model\"\n",
" pred_infected: dataframe with columns [\"total\", \"total_hos_min\", \"total_hos_max\", \"total_ICU_min\", \"total_ICU_max\"]\n",
" hospital_capacity: (int) number of hospital beds available. \n",
" plot_title: (str) title of plot \n",
" \"\"\"\n",
" first_dt = datetime.datetime.strptime(P[\"first_day\"], \"%m-%d-%Y\")\n",
" n_days = pred_infected.shape[0]\n",
"\n",
" t = list(range(0, n_days))\n",
" ax.plot(t, pred_infected[\"hos_max_total\"], label = 'liberal prediction: hospital beds', c = '#293EC9')\n",
" ax.plot(t, pred_infected[\"hos_min_total\"], label = 'conservative prediction: hospital beds', c = '#1FCED4')\n",
" ax.plot(t, pred_infected[\"ICU_max_total\"], label = 'liberal prediction: ICU beds', c= \"#119202\")\n",
" ax.plot(t, pred_infected[\"ICU_min_total\"], label ='conservative prediction: ICU beds', c = \"#EBC400\")\n",
" ax.hlines(hospital_capacity, t[0], t[-1], label = \"available hospital beds\", color = '#F95858')\n",
"\n",
" # label\n",
" time_label = [(first_dt + datetime.timedelta(days = i)).date().strftime('%m-%d') for i in range(0,n_days,20)]\n",
" ax.set_xticks(list(range(0,n_days,20))) \n",
" ax.set_xticklabels(time_label )\n",
" ax.set_xlabel(\"date\")\n",
" ax.set_ylabel(\"number of beds\")\n",
" ax.set_title(plot_title)\n",
"\n",
" return ax"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "F8n4MbtgM_y1",
"colab_type": "text"
},
"source": [
"# 1. Get data\n",
" * 1) Get COVID-19 data \n",
" * 2) Get US population info by state and age\n",
" * 3) Get hospital information by state"
]
},
{
"cell_type": "code",
"metadata": {
"id": "oViqR4FBdkTX",
"colab_type": "code",
"colab": {}
},
"source": [
"# mount google drive \n",
"from google.colab import drive\n",
"drive.mount('/content/drive')"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "wIMKS8UvmpMq",
"colab_type": "text"
},
"source": [
"## 1) Get COVID-19 data \n",
"* Grab data from [Johns Hopkins CSSE repository](https://github.com/CSSEGISandData/COVID-19). \n",
"* Grab data from 01/22/2020 to most recently updated data. \n",
"* **NOTE: While the following cell runs fine today (March 26, 2020), it could raise errors in the following situations:** \n",
" * If future data from CSSE is reported in a different format / header. \n",
" * If CSSE skips its daily update. "
]
},
{
"cell_type": "code",
"metadata": {
"id": "-A8nQZ7y1xWZ",
"colab_type": "code",
"colab": {}
},
"source": [
"url = \"https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_daily_reports/\"\n",
"day = datetime.datetime.strptime(\"01-22-2020\", \"%m-%d-%Y\").date()\n",
"\n",
"data = {}\n",
"error = False \n",
"\n",
"while error == False:\n",
" url_day = url + day.strftime(\"%m-%d-%Y\") + '.csv'\n",
" \n",
" # try grabbing data from given \"day\"\n",
" try:\n",
" df = pd.read_csv(url_day)\n",
"\n",
" # get rid of unnecessary columns (Latitutde, Longitude) if such columns exist\n",
" df.drop(columns = [\"Latitude\", \"Longitude\", \"Last Update\"], errors = \"ignore\", inplace = True)\n",
"\n",
" # create a column with report date, as given by date\n",
" df[\"report date\"] = [day.strftime(\"%m-%d-%Y\")] * df.shape[0]\n",
"\n",
" # Fill NaN values\n",
" values = {'Confirmed': 0, 'Deaths': 0, 'Recovered': 0}\n",
" df = df.fillna(value=values)\n",
" \n",
" # Make header uniform\n",
" if \"Country_Region\" in df.columns:\n",
" df.rename(columns={'Country_Region':'Country/Region'}, inplace=True)\n",
" if \"Province_State\" in df.columns:\n",
" df.rename(columns={'Province_State':'Province/State'}, inplace=True)\n",
" \n",
" # save to Dict\n",
" df = df[[\"Province/State\", \"Country/Region\", \"Confirmed\", \"Deaths\", \"Recovered\", \"report date\"]]\n",
" data[day.strftime(\"%m-%d-%Y\")] = df\n",
" \n",
" # update to next day\n",
" day = day + datetime.timedelta(days = 1)\n",
"\n",
" # if data from \"day\" doesn't exist\n",
" except HTTPError:\n",
" error = True \n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "S5hxegXcPgGQ",
"colab_type": "code",
"colab": {}
},
"source": [
"# save data\n",
"#directory = '/content/drive/My Drive/'\n",
"\n",
"#f = open(directory + \"COVID_data_03_26.pkl\",\"wb\")\n",
"#pickle.dump(data,f)\n",
"#f.close()"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "_t85kH-jl2Np",
"colab_type": "text"
},
"source": [
"## 2) Get US population info by state and age \n",
"* Download data from https://www.census.gov/data/tables/time-series/demo/popest/2010s-state-detail.html \n",
"* Download the data titled \"Annual Estimates of the Resident Population by Single Year of Age and Sex: April 1, 2010 to July 1, 2018 \". \n",
"* Save the \"PEP_2018_PEPSYASEX_with_ann.csv\" file in Google Drive "
]
},
{
"cell_type": "code",
"metadata": {
"id": "Z48mogcel64J",
"colab_type": "code",
"colab": {}
},
"source": [
"# load demograpahic data \n",
"dem_data = '/content/drive/My Drive/PEP_2018_PEPSYASEX/PEP_2018_PEPSYASEX_with_ann.csv'\n",
"US_dem = pd.read_csv(dem_data)\n",
"\n",
"# select columns that has the most recent (2018) population estimate of both sex\n",
"col = US_dem.columns[US_dem.columns.str.contains('2018', case = False ) \n",
" & US_dem.columns.str.contains('sex0', case = False )]\n",
"\n",
"# add state info\n",
"col = col.insert(0,\"GEO.display-label\")\n",
"US_dem = US_dem[col]\n",
"\n",
"US_dem = US_dem.drop([0])\n",
"\n",
"# rename the columns into something simpler\n",
"new_col = {x:re.sub('est72018sex0_', '',x) for x in col}\n",
"new_col[\"GEO.display-label\"] = \"GEO\"\n",
"\n",
"# change the names of the column \n",
"US_dem.rename(columns = new_col, inplace = True)\n",
"\n",
"# change type so that we can sum values\n",
"US_dem.loc[:,US_dem.columns != 'GEO'] = US_dem.loc[:,US_dem.columns != 'GEO'].astype(float)\n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "3yAQPtuvl7Eu",
"colab_type": "code",
"colab": {}
},
"source": [
"# Group ages according to \n",
"# https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e2.htm?s_cid=mm6912e2_w#T1_down\n",
"\n",
"# age 0-19\n",
"sum_col = [\"age\" + str(i) for i in range(0,20)]\n",
"US_dem[\"age0-19\"] = US_dem[sum_col].sum( axis = 1)\n",
" \n",
"# 20 -44\n",
"sum_col = [\"age\" + str(i) for i in range(20,45)]\n",
"US_dem[\"age20-44\"] = US_dem[sum_col].sum( axis = 1)\n",
"\n",
"# 45 - 54\n",
"sum_col = [\"age\"+str(i) for i in range(45,55)]\n",
"US_dem[\"age45-54\"] = US_dem[sum_col].sum( axis = 1)\n",
"\n",
"# 55 - 64\n",
"sum_col = [\"age\"+str(i) for i in range(55,65)]\n",
"US_dem[\"age55-64\"] = US_dem[sum_col].sum( axis = 1)\n",
"\n",
"# 65 - 74\n",
"sum_col = [\"age\"+str(i) for i in range(65,75)]\n",
"US_dem[\"age65-74\"] = US_dem[sum_col].sum( axis = 1)\n",
"\n",
"# 75 - 84\n",
"sum_col = [\"age\"+str(i) for i in range(75,85)]\n",
"US_dem[\"age75-84\"] = US_dem[sum_col].sum( axis = 1)\n",
"\n",
"# 85 <= already exists\n",
"US_dem = US_dem[['GEO', 'age999','age0-19','age20-44','age45-54','age55-64','age65-74','age75-84', 'age85plus']]\n",
"\n",
"# rewrite in terms of percentage \n",
"\n",
"US_dem[\"per0-19\"] = US_dem[\"age0-19\"] / US_dem[\"age999\"]\n",
"US_dem[\"per20-44\"] = US_dem[\"age20-44\"] / US_dem[\"age999\"]\n",
"US_dem[\"per45-54\"] = US_dem[\"age45-54\"] / US_dem[\"age999\"]\n",
"US_dem[\"per55-64\"] = US_dem[\"age55-64\"] / US_dem[\"age999\"]\n",
"US_dem[\"per65-74\"] = US_dem[\"age65-74\"] / US_dem[\"age999\"]\n",
"US_dem[\"per75-84\"] = US_dem[\"age75-84\"] / US_dem[\"age999\"]\n",
"US_dem[\"per85plus\"] = US_dem[\"age85plus\"] / US_dem[\"age999\"]"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "pINJgIqZgZsU",
"colab_type": "code",
"colab": {}
},
"source": [
"# save csv\n",
"#directory = '/content/drive/My Drive/'\n",
"#US_dem.to_csv(directory + \"US_demographics.csv\",index=False)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "GN2141u0NCAk",
"colab_type": "text"
},
"source": [
"## 3) Get hospital information (number of beds) by state\n",
"* data from https://www.ahd.com/state_statistics.html\n"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "DYtap3O9G2YJ",
"colab": {}
},
"source": [
"# create BeautifulSoup object\n",
"url = \"https://www.ahd.com/state_statistics.html\"\n",
"req = request.Request(\n",
" url, \n",
" data=None, \n",
" headers={\n",
" 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36'\n",
" }\n",
")\n",
"\n",
"html = urlopen(req)\n",
"soup = BeautifulSoup(html, \"lxml\") \n",
"\n",
"# table header\n",
"table_header = [x.text for x in soup.find_all('th')]\n",
"\n",
"# create df\n",
"US_hospitals = pd.DataFrame(columns = table_header)\n",
"\n",
"# find the table \n",
"table = soup.find('tbody')\n",
"\n",
"# find rows of table\n",
"table_rows = table.find_all('tr')\n",
"\n",
"for row in table_rows:\n",
" td = row.find_all('td')\n",
" row = [x.text for x in td]\n",
" row = [re.sub(',','',x) for x in row]\n",
"\n",
" row_df = pd.DataFrame(np.array([row]), columns = table_header)\n",
" US_hospitals = US_hospitals.append(row_df)\n",
"\n",
"US_hospitals.reset_index(inplace = True, drop = True)\n",
"US_hospitals[\"StaffedBeds\"] = pd.to_numeric(US_hospitals.StaffedBeds)\n",
"\n",
"US_hospitals = US_hospitals[[\"State\", \"NumberHospitals\", \"StaffedBeds\"]]"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "faCuQBI2UWOc",
"colab_type": "code",
"colab": {}
},
"source": [
"# save csv\n",
"#directory = '/content/drive/My Drive/'\n",
"#US_hospitals.to_csv(directory + \"US_hospitals.csv\",index=False)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "044dubCgscrh"
},
"source": [
"# 2. Model performance in Huebi, China \n",
"\n",
"* Use the data from first 22 days to build the model and make predictions \n",
"* Population in Hubei is about 58 million \n",
"* Model is built assuming that 0.13% of Hubei population will eventually contract the disease"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "w4IadDNVscrl",
"colab": {}
},
"source": [
"hubei_df = pd.DataFrame(columns = df.columns)\n",
"\n",
"for date in data.keys():\n",
" df = data[date]\n",
" hubei_data = df[((df['Country/Region'] == 'China') | (df[\"Country/Region\"] == \"Mainland China\"))\n",
" & (df[\"Province/State\"] == \"Hubei\") ]\n",
" hubei_df = hubei_df.append(hubei_data, ignore_index = True )\n",
"\n",
"# sort data, from oldest (top) to new (bottom) \n",
"hubei_df.sort_values(by = [\"report date\"], inplace = True)\n",
"hubei_df.reset_index(inplace = True)\n",
"\n",
"# adjust the number of infected people\n",
"hubei_df[\"Infected\"] = hubei_df[\"Confirmed\"] - hubei_df[\"Deaths\"] - hubei_df[\"Recovered\"]\n",
"hubei_df[\"death_or_recovered\"] = hubei_df[\"Deaths\"] + hubei_df[\"Recovered\"]"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "nKY4j6JxNqEr",
"colab_type": "code",
"outputId": "5606d083-95db-407f-a07b-c7518bdd4906",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 67
}
},
"source": [
"percentage = 0.0013\n",
"\n",
"# Population of Hubei, China\n",
"Hubei_population = 58000000\n",
"print(\"population: \", Hubei_population)\n",
"\n",
"# Susceptible population of state\n",
"n_susceptible = percentage * Hubei_population \n",
"print(\"number of susceptible population: \", int(n_susceptible ))\n",
"\n",
"print(\"As of March 23, 2020, the cumulative number of people confirmed in Hubei, China are 67,800\")\n",
"# For comparison, as of March 23 2020, the cumulative number of people confirmed\n",
"# in Hubei, China are 67,800. \n",
"# Let's use https://www.statista.com/statistics/1090007/china-confirmed-and-suspected-wuhan-coronavirus-cases-region/"
],
"execution_count": 26,
"outputs": [
{
"output_type": "stream",
"text": [
"population: 58000000\n",
"number of susceptible population: 75400\n",
"As of March 23, 2020, the cumulative number of people confirmed in Hubei, China are 67,800\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "1kuFA96xOi_s",
"colab_type": "code",
"colab": {}
},
"source": [
"# use the first X days of data to train the model \n",
"data_n_days = 22\n",
"\n",
"first_day = hubei_df[\"report date\"].values[0]\n",
"last_day = hubei_df[\"report date\"].values[data_n_days]"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "7DRYaIBUN2VC",
"colab_type": "code",
"colab": {}
},
"source": [
"P = fit_model(\"Hubei\", hubei_df, first_day, last_day, n_susceptible, 100)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "pv1HUuc5O-qg",
"colab_type": "code",
"outputId": "ae5d2c5c-20a4-4ce2-f021-76fc91a94393",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 422
}
},
"source": [
"fig, ax = plt.subplots(ncols = 2, figsize = (15,5))\n",
"plot_pred(ax[0], P, hubei_df, \"Infection prediction in Hubei, China for 100 days\")\n",
"\n",
"# zoom in on first 30 days\n",
"plot_pred(ax[1], P, hubei_df, \"Infection prediction in Hubei, China for immediate future\")\n",
"ax[1].set_xlim(0, 50)\n",
"ax[1].set_ylim(0,60000)\n",
"\n",
"# plot the data that were not used in the model fitting process\n",
"unused_data = hubei_df[hubei_df[\"report date\"] > last_day]\n",
"n_unused = unused_data.shape[0]\n",
"ax[0].scatter(list(range(data_n_days, data_n_days + n_unused )),unused_data[\"Infected\"].values, \n",
" marker = 'x', c = '#FF007F', label = 'data: infected')\n",
"ax[0].scatter(list(range(data_n_days, data_n_days + n_unused )), unused_data[\"death_or_recovered\"].values, \n",
" marker = 'x', c = '#00CCCC', label = \"data: deceased or recovered\")\n",
"\n",
"ax[1].scatter(list(range(data_n_days, data_n_days + n_unused )),unused_data[\"Infected\"].values, \n",
" marker = 'x', c = '#FF007F', label = 'new data: infected')\n",
"ax[1].scatter(list(range(data_n_days, data_n_days + n_unused )), unused_data[\"death_or_recovered\"].values, \n",
" marker = 'x', c = '#00CCCC', label = \"new data: deceased or recovered\")\n",
"\n",
"ax.flatten()[1].legend(loc='upper center', bbox_to_anchor=(-0.1, -0.12), ncol=2)"
],
"execution_count": 29,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.legend.Legend at 0x7fc05f3f44e0>"
]
},
"metadata": {
"tags": []
},
"execution_count": 29
},
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA5gAAAGECAYAAABTQ490AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<Figure size 1080x360 with 2 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0NxnHahHLKMa",
"colab_type": "text"
},
"source": [
"# 3. Analysis by state\n",
"\n",
"(1) Select state and collect state-level data \n",
"(2) Check if the state has enough data \n",
" * If the state has insufficient report on confirmed cases\n",
" * If the state has enough data on confimred cases, but does not have enough report on recovered data, simulate some recovered data before fitting model. \n",
"\n",
"(3) Select percentage of population that are susceptible to COVID-19 (USER INPUT REQUIRED) \n",
"(4) Predict number of infections, deaths, and recoveries. \n",
"(5) Predict hospital and ICU admission of state. \n",
"(6) Select percentage of hospital beds available (USER INPUT REQUIRED) \n",
"(7) Predict dates of hospital bed shortage and number of extra beds needed. \n",
"\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "tVpm57x2Bvyx",
"colab_type": "text"
},
"source": [
"(1) Select state, get state-level data"
]
},
{
"cell_type": "code",
"metadata": {
"id": "2qBqOM0wBxs3",
"colab_type": "code",
"outputId": "b879a32d-18ce-4a77-bfa1-ea2e5e7df89c",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 84
}
},
"source": [
"# select state using state abbreviations\n",
"state = \"NY\"\n",
"\n",
"# Collect state-level data \n",
"state_name = abbreviation[state]\n",
"state_df = get_state_df(data, state, state_name)\n",
"\n",
"# Print most recent status\n",
"date = state_df.tail(1)[\"report date\"].values[0]\n",
"confirmed = state_df.tail(1)[\"Confirmed\"].values[0]\n",
"death = state_df.tail(1)[\"Deaths\"].values[0]\n",
"\n",
"print(state + \" data as of \", date, \":\")\n",
"print(\"Confirmed cases: \", confirmed)\n",
"print(\"Number of death: \", death)"
],
"execution_count": 106,
"outputs": [
{
"output_type": "stream",
"text": [
"ERROR! Session/line number was not unique in database. History logging moved to new session 66\n",
"NY data as of 03-26-2020 :\n",
"Confirmed cases: 37877\n",
"Number of death: 385\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "fSzcuqBcGgyb",
"colab_type": "text"
},
"source": [
"(2) Check if state has enough data"
]
},
{
"cell_type": "code",
"metadata": {
"id": "QKg6ft_bGgTm",
"colab_type": "code",
"colab": {}
},
"source": [
"sufficient, first_day, last_day = check_data_sufficiency(state_df)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "HF1ztCsh5AgM",
"colab_type": "text"
},
"source": [
"(3) ***USER INPUT REQUIRED***: Select susceptible population \n",
"* Select percentage of state population that are expected to contract COVID-19 \n",
"* As a reference, about 0.12% of the population tested positive in Hubei, China \n",
"* As of 03/26/2020, about 0.19% of the State of New York population have tested positive. "
]
},
{
"cell_type": "code",
"metadata": {
"id": "kSQsfP4WRIvt",
"colab_type": "code",
"outputId": "4864fb06-5c82-46cd-a020-3e60a5ea6965",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 50
}
},
"source": [
"percentage = 0.01\n",
"\n",
"# Population of state\n",
"state_population = US_dem[US_dem[\"GEO\"] == state_name][\"age999\"].values[0]\n",
"print(\"state population: \", state_population)\n",
"\n",
"# Susceptible population of state\n",
"n_susceptible = percentage * state_population \n",
"print(\"number of susceptible population: \", n_susceptible )\n"
],
"execution_count": 108,
"outputs": [
{
"output_type": "stream",
"text": [
"state population: 19542209.0\n",
"number of susceptible population: 195422.09\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "_rSbY4MRG1y1",
"colab_type": "text"
},
"source": [
"(4) Predict number of infections, deaths, and recovery"
]
},
{
"cell_type": "code",
"metadata": {
"id": "p2HkbD47jY6S",
"colab_type": "code",
"colab": {}
},
"source": [
"if sufficient == False:\n",
" print(\"The selected state has insufficient data for predictions.\")\n",
"if sufficient == True:\n",
" P = fit_model(state, state_df, first_day, last_day, n_susceptible, 100)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "qgcsEQDdtbZi",
"colab_type": "code",
"outputId": "b7bc65cc-b4bb-4f15-f50f-3a138ea045f5",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 407
}
},
"source": [
"fig, ax = plt.subplots(ncols = 2, figsize = (15,5))\n",
"plot_pred(ax[0], P, state_df, \"Infection prediction in \" + str(state) + \" for 100 days\")\n",
"\n",
"# zoom in on first 30 days\n",
"plot_pred(ax[1], P, state_df, \"Infection prediction in \" + str(state) + \" for immediate future\")\n",
"ax[1].set_xlim(0, 40)\n",
"ax[1].set_ylim(0,state_df.tail(1)[\"Infected\"].values[0] * 1.5)\n",
"ax.flatten()[1].legend(loc='upper center', bbox_to_anchor=(-0.1, -0.12), ncol=2)"
],
"execution_count": 110,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.legend.Legend at 0x7fc056acbda0>"
]
},
"metadata": {
"tags": []
},
"execution_count": 110
},
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA58AAAF1CAYAAABvUV6xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwScREREREREtOQafREREREREtOT+AK5zUdptl/OWAAAAAElFTkSuQmCC\n",
"text/plain": [
"<Figure size 1080x360 with 2 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "gbD2AilHJsOg",
"colab_type": "text"
},
"source": [
"NOTE: If the selected state doesn't have enough data on COVID-19, the model may be completely off. In such cases, the graph above will do a poor job tracing the data. If the number of \"Confirmed\" and \"Deaths\" are too low, then the model will do a poor job making its predictions. "
]
},
{
"cell_type": "code",
"metadata": {
"id": "J0rf5IweM3Kg",
"colab_type": "code",
"outputId": "c9fc1944-87b4-4b3b-fdef-b9c24888078a",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 195
}
},
"source": [
"state_df.tail()"
],
"execution_count": 111,
"outputs": [
{
"output_type": "execute_result",
"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>Province/State</th>\n",
" <th>Country/Region</th>\n",
" <th>Confirmed</th>\n",
" <th>Deaths</th>\n",
" <th>Recovered</th>\n",
" <th>report date</th>\n",
" <th>Infected</th>\n",
" <th>death_or_recovered</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>20</th>\n",
" <td>New York</td>\n",
" <td>US</td>\n",
" <td>15800</td>\n",
" <td>117</td>\n",
" <td>0</td>\n",
" <td>03-22-2020</td>\n",
" <td>15157</td>\n",
" <td>643</td>\n",
" </tr>\n",
" <tr>\n",
" <th>21</th>\n",
" <td>New York</td>\n",
" <td>US</td>\n",
" <td>20884</td>\n",
" <td>158</td>\n",
" <td>0</td>\n",
" <td>03-23-2020</td>\n",
" <td>19938</td>\n",
" <td>946</td>\n",
" </tr>\n",
" <tr>\n",
" <th>22</th>\n",
" <td>New York</td>\n",
" <td>US</td>\n",
" <td>25681</td>\n",
" <td>210</td>\n",
" <td>0</td>\n",
" <td>03-24-2020</td>\n",
" <td>24336</td>\n",
" <td>1345</td>\n",
" </tr>\n",
" <tr>\n",
" <th>23</th>\n",
" <td>New York</td>\n",
" <td>US</td>\n",
" <td>30841</td>\n",
" <td>285</td>\n",
" <td>0</td>\n",
" <td>03-25-2020</td>\n",
" <td>29010</td>\n",
" <td>1831</td>\n",
" </tr>\n",
" <tr>\n",
" <th>24</th>\n",
" <td>New York</td>\n",
" <td>US</td>\n",
" <td>37877</td>\n",
" <td>385</td>\n",
" <td>0</td>\n",
" <td>03-26-2020</td>\n",
" <td>35465</td>\n",
" <td>2412</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Province/State Country/Region ... Infected death_or_recovered\n",
"20 New York US ... 15157 643\n",
"21 New York US ... 19938 946\n",
"22 New York US ... 24336 1345\n",
"23 New York US ... 29010 1831\n",
"24 New York US ... 35465 2412\n",
"\n",
"[5 rows x 8 columns]"
]
},
"metadata": {
"tags": []
},
"execution_count": 111
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "QWwbvMSbHFlK",
"colab_type": "text"
},
"source": [
"(5) Predict hospital and ICU admission "
]
},
{
"cell_type": "code",
"metadata": {
"id": "6uWmNJgyldRF",
"colab_type": "code",
"colab": {}
},
"source": [
"# get state demographics\n",
"state_dem = US_dem[US_dem[\"GEO\"] == state_name]\n",
"\n",
"# predict hospitalization and ICU admission\n",
"hos_prediction = predict_hospitalization_ICU(state_dem, P)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "PJamHo4HIiY9",
"colab_type": "text"
},
"source": [
"(6) ***USER INPUT REQUIRED***: Select percentage of hospital beds available. \n",
"* On average, about 35% of hospital beds are available for new patients. "
]
},
{
"cell_type": "code",
"metadata": {
"id": "9nIrR04-AFA0",
"colab_type": "code",
"outputId": "d8aabc97-14ab-4da4-8c88-9e734024466c",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 50
}
},
"source": [
"# select available bed percentage.\n",
"# must be between 0 and 1 \n",
"bed_percentage = 0.35\n",
"\n",
"# compare to state-wide hospital capacity\n",
"state_beds = int(US_hospitals[US_hospitals[\"State\"].str.contains(state)][\"StaffedBeds\"].values[0])\n",
"state_available = state_beds * bed_percentage\n",
"\n",
"print(\"number of hospital beds in \"+ state +\": \" , state_beds)\n",
"print(\"number of available beds in \" + state + \": \", int(state_available))"
],
"execution_count": 113,
"outputs": [
{
"output_type": "stream",
"text": [
"number of hospital beds in NY: 57261\n",
"number of available beds in NY: 20041\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WR8C8J9HJCyF",
"colab_type": "text"
},
"source": [
"(7) Predict hospital bed shortage"
]
},
{
"cell_type": "code",
"metadata": {
"id": "hQnD7d4JEYk-",
"colab_type": "code",
"outputId": "c81887ac-0a95-458c-eab4-2362abab8246",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 104
}
},
"source": [
"hospital_prediction(hos_prediction, state_available, state)"
],
"execution_count": 114,
"outputs": [
{
"output_type": "stream",
"text": [
"According to a conservative estimate, hospitals in NY will need more beds from 04-02-2020 to 04-13-2020\n",
"According to a conservative estimate, hospitals in NY will experience the highest level of bed shortage on 04-06-2020, and they will need 2921 extra beds.\n",
"According to a liberal estimate, hospitals in NY will need more beds from 03-29-2020 to 04-29-2020\n",
"According to a liberal estimate, hospitals in NY will experience the highest level of bed shortage on 04-06-2020, and they will need 15124 extra beds.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "PJkqDifLm-8g",
"colab_type": "code",
"outputId": "6c04f4c3-db55-4c4d-8359-f08d29508873",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 422
}
},
"source": [
"fig, ax = plt.subplots(ncols = 2, figsize = (15, 5))\n",
"plot_hospital_pred(ax[0], P, hos_prediction, state_available, 'hospital bed prediction in ' + str(state) +' for 100 days')\n",
"plot_hospital_pred(ax[1], P, hos_prediction, state_available, 'hospital bed prediction in ' + str(state) + ' for immediate future')\n",
"\n",
"ax[1].set_xlim(0, 50)\n",
"ax[1].set_ylim(0,state_available * 1.1)\n",
"ax.flatten()[1].legend(loc='upper center', bbox_to_anchor=(-0.1, -0.12), ncol=2)"
],
"execution_count": 116,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.legend.Legend at 0x7fc056954fd0>"
]
},
"metadata": {
"tags": []
},
"execution_count": 116
},
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA4oAAAGECAYAAABwGN6hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCCGEEEIIYUQCRSGEEEIIIYQQRv4HtCHJmEST1DoAAAAASUVORK5CYII=\n",
"text/plain": [
"<Figure size 1080x360 with 2 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "cGHKcp4XfnWo",
"colab_type": "text"
},
"source": [
"# 4. Analysis by county (in State of Washington) \n",
"(1) Get hospital information by county in WA \n",
"(2) Make hospital admission prediction on WA as a state \n",
"(3) Get demographic information of WA by county \n",
"(4) Select county and make hospital bed shortage prediction for selected county"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "l3QBLuWL39dL",
"colab_type": "text"
},
"source": [
"(1) Get hospital information by county in WA \n",
"\n",
"* https://www.wsha.org/our-members/member-listing/"
]
},
{
"cell_type": "code",
"metadata": {
"id": "KsOD7kvugEJO",
"colab_type": "code",
"colab": {}
},
"source": [
"# create BeautifulSoup object\n",
"url = \"https://www.wsha.org/our-members/member-listing/\"\n",
"req = request.Request(\n",
" url, \n",
" data=None, \n",
" headers={\n",
" 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36'\n",
" }\n",
")\n",
"\n",
"html = urlopen(req)\n",
"soup = BeautifulSoup(html, \"lxml\") \n",
"\n",
"# find the table \n",
"table = soup.find('table', id = 'find-hospital-list')\n",
"\n",
"# find rows of table\n",
"table_rows = table.find_all('tr')\n",
"\n",
"# find table header \n",
"table_header = [x.text for x in table.find_all('th')]\n",
"\n",
"# create data frame\n",
"WA_hospitals = pd.DataFrame(columns = table_header)\n",
"\n",
"# add table data to data frame \n",
"for row in table_rows:\n",
" td = row.find_all('td')\n",
" row = [x.text for x in td]\n",
" if row != []:\n",
" row_df = pd.DataFrame(np.array([row]), columns = table_header)\n",
" WA_hospitals = WA_hospitals.append(row_df)\n",
"\n",
"# replace \"empty\" beds with 0\n",
"WA_hospitals[\"Beds\"] = WA_hospitals[\"Beds\"].replace('',0)\n",
"WA_hospitals[\"Beds\"] = WA_hospitals[\"Beds\"].astype(int)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "OOA5LyUYQiUo",
"colab": {}
},
"source": [
"# save csv\n",
"#directory = '/content/drive/My Drive/'\n",
"#WA_hospitals.to_csv(directory + \"WA_hospitals.csv\",index=False)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "PefkpkLeFK7D"
},
"source": [
"(2) Predict the number of cases in Washington state \n",
"* ***USER INPUT REQUIRED***: Select percentage of population that are susceptible to COVID-19. Percentage should be between 0 and 1"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"outputId": "f204433d-27ea-43b8-959f-e3ec207aaf70",
"id": "-WfxezAUFR6B",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 101
}
},
"source": [
"percentage = 0.01\n",
"state = \"WA\"\n",
"\n",
"# Collect state-level data \n",
"state_name = abbreviation[state]\n",
"state_df = get_state_df(data, state, state_name)\n",
"\n",
"# Print most recent status\n",
"date = state_df.tail(1)[\"report date\"].values[0]\n",
"confirmed = state_df.tail(1)[\"Confirmed\"].values[0]\n",
"death = state_df.tail(1)[\"Deaths\"].values[0]\n",
"\n",
"print(state + \" data as of \", date, \":\")\n",
"print(\"Confirmed cases: \", confirmed)\n",
"print(\"Number of death: \", death)\n",
"\n",
"# Find the first and last day used for fitting the model\n",
"sufficient, first_day, last_day = check_data_sufficiency(state_df)\n",
"\n",
"# Population of state\n",
"state_population = US_dem[US_dem[\"GEO\"] == state_name][\"age999\"].values[0]\n",
"print(\"state population: \", state_population)\n",
"\n",
"# Susceptible population of state\n",
"n_susceptible = percentage * state_population \n",
"print(\"number of susceptible population: \", n_susceptible )\n",
"\n",
"# Make prediction on infection\n",
"P = fit_model(state, state_df, first_day, last_day, n_susceptible, 100)\n",
"\n",
"# Make hospital and ICU admission prediction\n",
"state_dem = US_dem[US_dem[\"GEO\"] == state_name]\n",
"hos_prediction = predict_hospitalization_ICU(state_dem, P)"
],
"execution_count": 118,
"outputs": [
{
"output_type": "stream",
"text": [
"WA data as of 03-26-2020 :\n",
"Confirmed cases: 3207.0\n",
"Number of death: 150.0\n",
"state population: 7535591.0\n",
"number of susceptible population: 75355.91\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "oH9JZigzGXFk",
"colab_type": "text"
},
"source": [
"(3) Get demographic information by county"
]
},
{
"cell_type": "code",
"metadata": {
"id": "eq-F7lvIf4rW",
"colab_type": "code",
"colab": {}
},
"source": [
"options = webdriver.ChromeOptions()\n",
"options.add_argument('--headless')\n",
"options.add_argument('--no-sandbox')\n",
"options.add_argument('--disable-dev-shm-usage')\n",
"\n",
"# grab data from website\n",
"wd = webdriver.Chrome('chromedriver',options=options)\n",
"wd.get(\"https://www.doh.wa.gov/emergencies/coronavirus\")\n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "mnoxhWeVGn7I",
"colab_type": "code",
"colab": {}
},
"source": [
"wd.implicitly_wait(3)\n",
"# beautiful soup object\n",
"soup = BeautifulSoup(wd.page_source)\n",
"\n",
"# find table\n",
"for caption in soup.find_all('caption'):\n",
" \n",
" if \"by County\" in caption.get_text():\n",
" table = caption.find_parent('table', {'class': 'table table-striped'})\n",
"\n",
"table_rows = table.find_all('tr')\n",
"\n",
"# table header\n",
"WA_county_col = [x.text for x in table.find_all('th')]\n",
"\n",
"# save to DataFrame\n",
"WA_county = pd.DataFrame(columns = WA_county_col)\n",
"\n",
"for row in table_rows:\n",
" county_data = [x.text for x in row.find_all('td')]\n",
" if county_data != []:\n",
" county_df = pd.DataFrame(np.array([county_data]), columns = WA_county_col) \n",
" WA_county = WA_county.append(county_df)\n",
"\n",