{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# Import various Python packages.\n", "\n", "import pandas as pd\n", "import numpy as np\n", "import statistics\n", "import math\n", "import matplotlib.pyplot as plt\n", "import statsmodels.api as sm\n", "from statsmodels.graphics.gofplots import qqplot\n", "from scipy.stats import shapiro" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# Import the data set into a Python dataframe, which we will call df.\n", "# Clean the dataframe so we only include the columns CPT, Surgery Time, and Wait Time Target.\n", "\n", "df = pd.read_csv(\"On Time OR - Surgery durations (Richard Hoshino).csv\", \n", " names=[\"A\", \"B\", \"C\", \"CPT\", \"Time\", \"F\", \"G\", \"H\", \"I\", \"WTT\", \n", " \"K\", \"L\", \"M\", \"N\", \"O\", \"P\", \"Q\"])\n", "df = df.drop([\"A\", \"B\", \"C\", \"F\", \"G\", \"H\", \"I\", \"K\", \"L\", \"M\", \"N\", \"O\", \"P\", \"Q\"], axis=1)" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
CPTTimeWTTLogTime
011047154 hours2.708050
1110476024 hours4.094345
2110474024 hours3.688879
3110472024 hours2.995732
4110476024 hours4.094345
\n", "
" ], "text/plain": [ " CPT Time WTT LogTime\n", "0 11047 15 4 hours 2.708050\n", "1 11047 60 24 hours 4.094345\n", "2 11047 40 24 hours 3.688879\n", "3 11047 20 24 hours 2.995732\n", "4 11047 60 24 hours 4.094345" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Add a new column called LogTime, that takes the natural logarithm of the surgery time.\n", "# Then print out the first five rows of the new dataset.\n", "\n", "df[\"LogTime\"] = np.log(df[\"Time\"])\n", "df.head()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "There are 565 total surgeries in the dataset\n", "Time: mean and stdev are (53.32566371681416, 37.714157288181696)\n", "LogTime: mean and stdev are (3.7414152411397064, 0.721057550455386)\n" ] } ], "source": [ "# Determine the mean and standard deviation of the entire dataset, for both Time and LogTime.\n", "\n", "print(\"There are\", len(df), \"total surgeries in the dataset\")\n", "print(\"Time: mean and stdev are\", (np.mean(df[\"Time\"]), np.std(df[\"Time\"])))\n", "print(\"LogTime: mean and stdev are\", (np.mean(df[\"LogTime\"]), np.std(df[\"LogTime\"])))\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Actual 50th percentile: 50.0 minutes\n", "Actual 80th percentile: 75.0 minutes\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# We draw a Cumulative Distribution Function of the given data, to see the thresholds of surgery\n", "# times for each percentile. From the results below, we see that 50% of the surgeries last at \n", "# most 50 minutes, and 80% of the surgeries last at most 75 minutes.\n", "\n", "counts, bin_edges = np.histogram(df[\"Time\"], bins=50)\n", "cdf = np.cumsum (counts)\n", "plt.plot (bin_edges[1:], cdf/cdf[-1])\n", "\n", "for q in [50, 80]:\n", " print (\"Actual {}th percentile: {} minutes\".format (q, np.percentile(df[\"Time\"], q)))" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Predicted 50th percentile for Normal distribution: 53.33 minutes\n", "Predicted 80th percentile for Normal distribution: 85.07 minutes\n" ] } ], "source": [ "# If the Time variable is normally distributed, we would be able to forecast the expected\n", "# 50th and 80th percentiles, using the Z-values of Z=0 for p=0.5 and Z=0.8416 for p=0.8\n", "# Specifically, 50% of the data is at most mean + 0*stdev, and 80% of the data is at most\n", "# mean + 0.8416*stdev. We see that both calculations OVER-estimate the actual results of\n", "# 50 minutes and 75 minutes, respectively.\n", "\n", "percentile50 = np.mean(df[\"Time\"])\n", "percentile80 = np.mean(df[\"Time\"])+0.8416*np.std(df[\"Time\"])\n", "\n", "print(\"Predicted 50th percentile for Normal distribution:\", round(percentile50,2), \"minutes\")\n", "print(\"Predicted 80th percentile for Normal distribution:\", round(percentile80,2), \"minutes\")" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Predicted 50th percentile for LogNormal distribution: 42.16 minutes\n", "Predicted 80th percentile for LogNormal distribution: 77.34 minutes\n" ] } ], "source": [ "# If the Time variable is log-normally distributed, then that implies that the LogTime variable\n", "# is normally distributed. Thus, we re-run the same calculations as above. We see that this\n", "# prediction is much better, but still off from the actual results of 50 minutes and 75 minutes.\n", "\n", "percentile50 = math.exp(np.mean(df[\"LogTime\"]))\n", "percentile80 = math.exp(np.mean(df[\"LogTime\"])+0.8416*np.std(df[\"LogTime\"]))\n", "\n", "print(\"Predicted 50th percentile for LogNormal distribution:\", round(percentile50,2), \"minutes\")\n", "print(\"Predicted 80th percentile for LogNormal distribution:\", round(percentile80,2), \"minutes\")" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# We now plot the histograms for both Time and LogTime, dividing our data into 15 equal-sized\n", "# bins. From the picture, we see that the Time data is certainly not normally-distributed,\n", "# while the LogTime data is much closer to a normal distribution.\n", "\n", "plt.figure(figsize=(12,3)) \n", "plt.subplot(1,2,1)\n", "plt.hist(df[\"Time\"], bins=15)\n", "plt.xlabel(\"Time\")\n", "plt.ylabel(\"Frequency\")\n", "plt.subplot(1,2,2)\n", "plt.hist(df[\"LogTime\"], bins=15)\n", "plt.xlabel(\"LogTime\")\n", "plt.ylabel(\"Frequency\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# We use the Quantile-Quantile plot to see how normally distributed our variables are.\n", "# From the picture, we see that Time is nowhere close to being normally distributed, while \n", "# the LogTime data seems to fit a normal distribution, except at the endpoints.\n", "\n", "fig, ax = plt.subplots(1, 2, figsize=(12, 3))\n", "plot1 = sm.ProbPlot(df[\"Time\"], fit=True)\n", "plot1.qqplot(line='45', ax=ax[0])\n", "plot2 = sm.ProbPlot(df[\"LogTime\"], fit=True)\n", "plot2.qqplot(line='45', ax=ax[1])\n", "ax[0].set_title('Q-Q Plot of Time for All Surgeries')\n", "ax[1].set_title('Q-Q Plot of LogTime for All Surgeries')\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Shapiro-Wilk on Time: W = 0.847, p = 0.000\n", "Time data does not look Gaussian (reject H0)\n", "\n", "Shapiro-Wilk on LogTime: W = 0.977, p = 0.000\n", "LogTime data does not look Gaussian (reject H0)\n" ] } ], "source": [ "# Because there are so many samples (N=565) in our dataset, any distribution that isn't\n", "# normal will result in a extremely low p-value when using the Shapiro-Wilk test for Normality.\n", "# This is because the \"statistical power\" of the Shapiro-Wilk normality test is very high\n", "# when we have a large value of N. This explains why we have a p-value of 0.000 when applying\n", "# the Shapiro-Wilk test to both Time and LogTime.\n", "\n", "stat, p = shapiro(df[\"Time\"])\n", "print(\"Shapiro-Wilk on Time:\", 'W = %.3f, p = %.3f' % (stat, p))\n", "if p > 0.05:\n", " print('Time data looks Gaussian (fail to reject H0)')\n", "else:\n", " print('Time data does not look Gaussian (reject H0)')\n", " print('')\n", "stat, p = shapiro(df[\"LogTime\"])\n", "print(\"Shapiro-Wilk on LogTime:\", 'W = %.3f, p = %.3f' % (stat, p))\n", "\n", "if p > 0.05:\n", " print('LogTime data looks Gaussian (fail to reject H0)')\n", "else:\n", " print('LogTime data does not look Gaussian (reject H0)')" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPT 25515 has 30 surgeries, with a Time p-value of 0.1 and a LogTime p-value of 0.476\n", "CPT 27506 has 38 surgeries, with a Time p-value of 0.0 and a LogTime p-value of 0.288\n", "CPT 27814 has 45 surgeries, with a Time p-value of 0.033 and a LogTime p-value of 0.224\n", "CPT 33217 has 28 surgeries, with a Time p-value of 0.0 and a LogTime p-value of 0.108\n", "CPT 44970 has 26 surgeries, with a Time p-value of 0.02 and a LogTime p-value of 0.435\n", "CPT 47562 has 27 surgeries, with a Time p-value of 0.86 and a LogTime p-value of 0.036\n", "CPT 52353 has 59 surgeries, with a Time p-value of 0.0 and a LogTime p-value of 0.009\n", "CPT 59618 has 58 surgeries, with a Time p-value of 0.1 and a LogTime p-value of 0.05\n" ] } ], "source": [ "# What we will do then is split our dataset by surgery type, using the CPT code.\n", "# Specifically, we calculate the Shapiro-Wilk test for each CPT code, provided we have\n", "# at least 25 surgeries in the group. We determine the p-values for the Shapiro-Wilk test,\n", "# for both Time and LogTime data.\n", "\n", "for i in df[\"CPT\"].unique():\n", " count=0\n", " for j in range(len(df)):\n", " if df[\"CPT\"][j]==i:\n", " count+=1\n", " if count >= 25:\n", " filter = (df[\"CPT\"] == i)\n", " newdata = df[filter]\n", " stat, p1 = shapiro(newdata[\"Time\"])\n", " stat, p2 = shapiro(newdata[\"LogTime\"])\n", " if p1<10 and p2>0:\n", " print(\"CPT\", i, \"has\", count, \"surgeries, with a Time p-value of\", \n", " round(p1,3), \"and a LogTime p-value of\", round(p2,3))" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# CPT=27506 (Femural Nail) appears to have a LogNormal distribution, given the low p-value (0.0)\n", "# for Time, and the high p-value (0.288) for LogTime. Let's check the Quantile-Quantile plots.\n", "\n", "filter = (df[\"CPT\"] == 27506)\n", "newdata = df[filter]\n", "fig, ax = plt.subplots(1, 2, figsize=(12, 3))\n", "plot1 = sm.ProbPlot(newdata[\"Time\"], fit=True)\n", "plot1.qqplot(line='45', ax=ax[0])\n", "plot2 = sm.ProbPlot(newdata[\"LogTime\"], fit=True)\n", "plot2.qqplot(line='45', ax=ax[1])\n", "ax[0].set_title('Q-Q Plot of Time for CPT=27506')\n", "ax[1].set_title('Q-Q Plot of LogTime for CPT=27506')\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# CPT=47562 (Laparotomy) appears to have a Normal distribution, given the high p-value (0.86)\n", "# for Time, and the low p-value (0.036) for LogTime. Let's check the Quantile-Quantile plots.\n", "\n", "filter = (df[\"CPT\"] == 47562)\n", "newdata = df[filter]\n", "fig, ax = plt.subplots(1, 2, figsize=(12, 3))\n", "plot1 = sm.ProbPlot(newdata[\"Time\"], fit=True)\n", "plot1.qqplot(line='45', ax=ax[0])\n", "plot2 = sm.ProbPlot(newdata[\"LogTime\"], fit=True)\n", "plot2.qqplot(line='45', ax=ax[1])\n", "ax[0].set_title('Q-Q Plot of Time for CPT=47562')\n", "ax[1].set_title('Q-Q Plot of LogTime for CPT=47562')\n", "plt.show()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.3" } }, "nbformat": 4, "nbformat_minor": 2 }