Rheology_Analysis_Toolbox_Notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "981aac81",
   "metadata": {},
   "source": [
    "# Source Code - Rheological Analysis Toolbox Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "30f4f39b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import openpyxl\n",
    "import datetime\n",
    "import math\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as mticker\n",
    "import seaborn as sns\n",
    "import plotly.graph_objects as go\n",
    "from ipywidgets import interact, FloatRangeSlider\n",
    "from prettytable import PrettyTable\n",
    "from fpdf import FPDF\n",
    "import re\n",
    "import scipy\n",
    "from scipy import stats as stats\n",
    "from scipy.stats import t\n",
    "from scipy.stats import ttest_ind\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "\n",
    "\n",
    "\n",
    "def create_new_workbook(file_name, column_labels):\n",
    "    \"\"\"\n",
    "    Create a new workbook with the given file name and add column_labels as the header.\n",
    "\n",
    "    Parameters:\n",
    "        file_name (str): Name of the Excel file to be created.\n",
    "        column_labels (list): List of strings representing the header labels.\n",
    "\n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"\n",
    "    wb = openpyxl.Workbook()\n",
    "    ws = wb.active\n",
    "    ws.append(column_labels)\n",
    "    wb.save(file_name)\n",
    "\n",
    "    \n",
    "def get_unique_file_name(file_name):\n",
    "    \"\"\"\n",
    "    Add a number to the file name to make it unique.\n",
    "\n",
    "    Parameters:\n",
    "        file_name (str): Name of the Excel file.\n",
    "\n",
    "    Returns:\n",
    "        str: Unique file name with a number appended to it.\n",
    "    \"\"\"\n",
    "    file_name_parts = file_name.split('.')\n",
    "    base_name = file_name_parts[0]\n",
    "    extension = file_name_parts[1]\n",
    "    counter = 1\n",
    "\n",
    "    unique_file_name = f\"{base_name}_{counter}.{extension}\"\n",
    "    \n",
    "    # Check if the file with the generated unique name already exists\n",
    "    while os.path.exists(unique_file_name):\n",
    "        counter += 1\n",
    "        unique_file_name = f\"{base_name}_{counter}.{extension}\"\n",
    "\n",
    "    return unique_file_name\n",
    "\n",
    "\n",
    "def create_database(file_name, column_labels):\n",
    "    \"\"\"\n",
    "    Create a new database as an empty Excel file with specified headers.\n",
    "\n",
    "    Parameters:\n",
    "        file_name (str): Name of the Excel file to be created.\n",
    "        column_labels (str): Column labels for the database\n",
    "    \n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"\n",
    "    # Check if the file already exists\n",
    "    if os.path.exists(file_name):\n",
    "        # Prompt the user to choose whether to replace or add to the file\n",
    "        choice = input('A file with this name already exists. Do you want to (R)eplace or (A)dd to the file? ').lower()\n",
    "        if choice == 'r':\n",
    "            # Create a new workbook and save it with the same file name\n",
    "            create_new_workbook(file_name, column_labels)\n",
    "        elif choice == 'a':\n",
    "            # Add a number to the file name to make it unique\n",
    "            unique_file_name = get_unique_file_name(file_name)\n",
    "            create_new_workbook(unique_file_name, column_labels)\n",
    "        else:\n",
    "            print('Invalid choice. Exiting...')\n",
    "            exit()\n",
    "    else:\n",
    "        # Create a new workbook and save it with the file name\n",
    "        create_new_workbook(file_name, column_labels)\n",
    "\n",
    "        \n",
    "        \n",
    "def get_valid_integer(prompt):\n",
    "    \"\"\"\n",
    "    Prompt the user to input a valid integer.\n",
    "\n",
    "    Parameters:\n",
    "        prompt (str): The prompt message for the user.\n",
    "\n",
    "    Returns:\n",
    "        int: Valid integer provided by the user.\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        user_input = input(prompt).strip()\n",
    "        if not user_input.isdigit():\n",
    "            print(\"Invalid input. Please enter a valid integer.\")\n",
    "        else:\n",
    "            return int(user_input)\n",
    "\n",
    "        \n",
    "def get_valid_float(prompt):\n",
    "    \"\"\"\n",
    "    Prompt the user to input a valid float.\n",
    "\n",
    "    Parameters:\n",
    "        prompt (str): The prompt message for the user.\n",
    "\n",
    "    Returns:\n",
    "        float: Valid float provided by the user.\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        user_input = input(prompt).strip()\n",
    "        try:\n",
    "            number = float(user_input)\n",
    "            return number\n",
    "        except ValueError:\n",
    "            print(\"Invalid input. Please enter a valid float.\")\n",
    "\n",
    "            \n",
    "def get_valid_date():\n",
    "    \"\"\"\n",
    "    Prompt the user to input a valid date in the format (DD/MM/YYYY).\n",
    "    \n",
    "    Parameters: \n",
    "        None.\n",
    "    \n",
    "    Returns:\n",
    "        str: A valid date in the format (DD/MM/YYYY).\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        date_input = input(\"Date: (DD/MM/YYYY)\").strip()\n",
    "        try:\n",
    "            date_obj = datetime.datetime.strptime(date_input, \"%d/%m/%Y\")\n",
    "            return date_input\n",
    "        except ValueError:\n",
    "            print(\"Invalid date format. Please enter the date in the format (DD/MM/YYYY).\")\n",
    "\n",
    "            \n",
    "def get_valid_input(prompt, valid_values=None):\n",
    "    \"\"\"\n",
    "    Prompt the user to input valid data based on the provided prompt and allowed values.\n",
    "\n",
    "    Parameters:\n",
    "        prompt (str): The prompt message for the user.\n",
    "        valid_values (list, optional): A list of valid values (strings) if applicable. Default value: None.\n",
    "\n",
    "    Returns:\n",
    "        str: Valid input provided by the user.\n",
    "    \"\"\"\n",
    "    while True:\n",
    "        user_input = input(prompt).strip().lower()\n",
    "\n",
    "        if valid_values is not None and user_input not in valid_values:\n",
    "            valid_values_str = \", \".join(valid_values)\n",
    "            print(f\"Invalid input. Please choose from: {valid_values_str}\")\n",
    "        elif user_input == \"\":\n",
    "            print(\"Empty input not allowed. Please provide a value.\")\n",
    "        else:\n",
    "            return user_input\n",
    "\n",
    "        \n",
    "def sdescription():\n",
    "    \"\"\"\n",
    "    Collect and process sample data from user input.\n",
    "    \n",
    "    Parameters:\n",
    "        None.\n",
    "    Returns:\n",
    "        list: A list containing sample information in the following order:\n",
    "        [Sample ID, Date, Gender, Age, Joint, Condition, Total Volume (mL), Color,\n",
    "        Transparency, Texture, Blood, Clot, Tissue, Storage in freezer, \n",
    "        COVID-19, Vaccinated for COVID-19, COVID-19 now, Number of Tests]\n",
    "    \"\"\"\n",
    "    sdesc = []\n",
    "\n",
    "    # Prompt the user to input data. Collect sample attributes\n",
    "    snum = get_valid_integer(\"Sample's number: \")\n",
    "    sid = 'S' + str(snum)\n",
    "    sdesc.append(sid)\n",
    "\n",
    "    sdate = get_valid_date()\n",
    "    sdesc.append(sdate)\n",
    "\n",
    "    sgender = get_valid_input(\"Gender: (male, female)\", valid_values=[\"male\", \"female\"])\n",
    "    sdesc.append(sgender)\n",
    "\n",
    "    sage = get_valid_integer(\"Age: \")\n",
    "    sdesc.append(sage)\n",
    "\n",
    "    sjoint = get_valid_input(\"Joint: (knee, hip, elbow, wrist, etc.)\", \n",
    "                             valid_values=[\"knee\", \"hip\", \"elbow\", \"wrist\"])\n",
    "    sdesc.append(sjoint)\n",
    "\n",
    "    scon = get_valid_input(\"Condition: (oa, healthy, meniscal tear, etc.)\", \n",
    "                           valid_values=[\"oa\", \"healthy\", \"meniscal tear\", \"patellofemoral pain\", \"femoral head necrosis\", \n",
    "                                        \"shoulder instability\", \"acl rupture\", \"oa broken poly\", \"nailing hip\", \n",
    "                                         \"patellofemoral cartilage defect\"])\n",
    "    sdesc.append(scon)\n",
    "    \n",
    "    svol = get_valid_float(\"Total volume (mL): '\")\n",
    "    sdesc.append(svol)\n",
    "\n",
    "    scolor = get_valid_input(\"Color: (red, orange, yellow, soft yellow, colorless)\", \n",
    "                             valid_values=[\"red\", \"orange\", \"yellow\", \"soft yellow\", \"colorless\"])\n",
    "    sdesc.append(scolor)\n",
    "    \n",
    "    strans = get_valid_input(\"Transparency: (transparent, non-transparent)\", \n",
    "                              valid_values=[\"transparent\", \"non-transparent\"])\n",
    "    sdesc.append(strans)\n",
    "    \n",
    "    stexture = get_valid_input(\"Texture: (viscous, non-viscous)\", valid_values=[\"viscous\", \"non-viscous\"])\n",
    "    sdesc.append(stexture)\n",
    "    \n",
    "    sblood = get_valid_input(\"Blood: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(sblood)\n",
    "\n",
    "    sclot = get_valid_input(\"Clot: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(sclot)\n",
    "\n",
    "    stissue = get_valid_input(\"Tissue: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(stissue)\n",
    "    \n",
    "    frstorage = get_valid_input(\"Storage in freezer (-20oC): (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(frstorage)\n",
    "\n",
    "    covid = get_valid_input(\"COVID-19: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(covid)\n",
    "\n",
    "    vaccine = get_valid_input(\"Vaccinated for COVID-19: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(vaccine)\n",
    "\n",
    "    covid_now = get_valid_input(\"COVID-19 now: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "    sdesc.append(covid_now)\n",
    "    \n",
    "    tnum = get_valid_integer(\"Number of test: \")\n",
    "    sdesc.append(tnum)\n",
    "\n",
    "    # Print the input data for review\n",
    "    print('\\nInput data:')\n",
    "    for idx, item in enumerate(sdesc, start=1):\n",
    "        print(f'{idx}: {item}')\n",
    "\n",
    "    # Prompt the user to confirm or revise the input data\n",
    "    choice = input('\\nDo you want to (S)ave the data or (R)evise the data? ').lower()\n",
    "    if choice == 's':\n",
    "        print('Sample data are saved.')\n",
    "        return sdesc\n",
    "    elif choice == 'r':\n",
    "        # Allow the user to go back and revise any input point\n",
    "        while True:\n",
    "            revise = input('\\nWhich data point do you want to revise? (1, 2, ... 18) or (E)xit to save: ').lower()\n",
    "            if revise == 'e':\n",
    "                break\n",
    "            elif revise.isdigit() and 1 <= int(revise) <= 18:\n",
    "                revise_idx = int(revise) - 1\n",
    "                revised_value = input(f'Revise {sdesc[revise_idx]}: ')\n",
    "                sdesc[revise_idx] = revised_value\n",
    "            else:\n",
    "                print('Invalid choice.')\n",
    "\n",
    "            # Print the revised input data for review\n",
    "            print('\\nRevised input data:')\n",
    "            for idx, item in enumerate(sdesc, start=1):\n",
    "                print(f'{idx}: {item}')\n",
    "\n",
    "            # Prompt the user to confirm or revise the input data again\n",
    "            choice = input('\\nDo you want to (S)ave the data or (R)evise the data again? ').lower()\n",
    "            if choice == 's':\n",
    "                print('Sample data are saved.')\n",
    "                return sdesc\n",
    "            elif choice == 'r':\n",
    "                continue\n",
    "        # Add a return statement to ensure the function always returns a list\n",
    "        print('There might be a problem. Check the Excel Workbook.')\n",
    "        return sdesc\n",
    "    else:\n",
    "        # Add a return statement to ensure the function always returns a list\n",
    "        print('There might be a problem. Check the Excel Workbook.')\n",
    "        exit()\n",
    "\n",
    "        \n",
    "def tdescription(my_list):\n",
    "    \"\"\"\n",
    "    Collect and process test data from user input.\n",
    "    \n",
    "    Parameters:\n",
    "        list: A list containing sample information in the following order:\n",
    "        [Sample ID, Date, Gender, Age, Joint, Condition, Total Volume (mL), \n",
    "        Color, Transparency, Texture, Blood, Clot, Tissue, Storage in freezer, \n",
    "        COVID-19, Vaccinated for COVID-19, COVID-19 now, Number of Tests]\n",
    "        \n",
    "    Returns:\n",
    "        list: A list containing sample information in the following order:\n",
    "        [Test Type, Temperature (oC), Tissue in Test, Bubbles before Test, Bubbles after Test,]\n",
    "    \"\"\"\n",
    "    stissue = my_list[12]\n",
    "    tnum = int(my_list[17])\n",
    "    \n",
    "    tdesc = []\n",
    "    \n",
    "    for i in range(tnum):\n",
    "        # Prompt the user to input data. Collect test attributes\n",
    "        ttype = get_valid_input(\"Test type: (time sweep, frequency sweep, flow step, strain sweep)\", \n",
    "                                  valid_values=[\"time sweep\", \"frequency sweep\", \"flow step\", \"strain sweep\"])\n",
    "\n",
    "        ttemp = get_valid_input(\"Temperature (oC): (25, 37)\", valid_values=[\"25\", \"37\"])\n",
    "\n",
    "        if stissue == 'yes':\n",
    "            ttissue = get_valid_input(\"Tissue in test: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "        else:\n",
    "            ttissue = 'no'\n",
    "\n",
    "        btbubbles = get_valid_input(\"Bubbles before test: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "\n",
    "        atbubbles = get_valid_input(\"Bubbles after test: (yes, no)\", valid_values=[\"yes\", \"no\"])\n",
    "        \n",
    "        test_info = [ttype, ttemp, ttissue, btbubbles, atbubbles]\n",
    "        \n",
    "        # Print the input data for review\n",
    "        print('\\nInput data:')\n",
    "        for idx, item in enumerate(test_info, start=1):\n",
    "            print(f'{idx}: {item}')\n",
    "\n",
    "        # Prompt the user to confirm or revise the input data\n",
    "        choice = input('\\nDo you want to (S)ave the data or (R)evise the data? ').lower()\n",
    "        if choice == 's':\n",
    "            print('Test data are saved.')\n",
    "            tdesc.append(test_info)\n",
    "            \n",
    "        elif choice == 'r':\n",
    "            # Allow the user to go back and revise any input point\n",
    "            while True:\n",
    "                revise = input('\\nWhich data point do you want to revise? (1, 2, ... 5) or (E)xit to save: ').lower()\n",
    "                if revise == 'e':\n",
    "                    break\n",
    "                elif revise.isdigit() and 1 <= int(revise) <= 5:\n",
    "                    revise_idx = int(revise) - 1\n",
    "                    revised_value = input(f'Revise {test_info[revise_idx]}: ')\n",
    "                    test_info[revise_idx] = revised_value\n",
    "                else:\n",
    "                    print('Invalid choice.')\n",
    "\n",
    "                # Print the revised input data for review\n",
    "                print('\\nRevised input data:')\n",
    "                for idx, item in enumerate(test_info, start=1):\n",
    "                    print(f'{idx}: {item}')\n",
    "\n",
    "                # Prompt the user to confirm or revise the input data again\n",
    "                choice = input('\\nDo you want to (S)ave the data or (R)evise the data again? ').lower()\n",
    "                if choice == 's':\n",
    "                    print('Test data are saved.')\n",
    "                    tdesc.append(test_info)\n",
    "                    break\n",
    "                elif choice == 'r':\n",
    "                    continue\n",
    "                else:\n",
    "                    # Add a return statement to ensure the function always returns a list\n",
    "                    print('There might be a problem. Check the Excel Workbook.')\n",
    "                    break\n",
    "        else:\n",
    "            # Add a return statement to ensure the function always returns a list\n",
    "            print('There might be a problem. Check the Excel Workbook.')\n",
    "            exit()\n",
    "    return tdesc \n",
    "  \n",
    "    \n",
    "def append_sample_to_database(file_name=\"Data/Database.xlsx\"):\n",
    "    \"\"\"\n",
    "    Append new sample data to an existing Excel database.\n",
    "\n",
    "    Parameters:\n",
    "        file_name (str, optional): Name of the Excel file to be updated. Default value: \"Data/Database.xlsx\".\n",
    "\n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"\n",
    "    sample_data = sdescription()\n",
    "    test_data = tdescription(sample_data)\n",
    "\n",
    "    # Load the existing Excel file\n",
    "    wb = openpyxl.load_workbook(file_name)\n",
    "    sheet = wb.active\n",
    "\n",
    "    # Find the next empty row in the Excel sheet\n",
    "    next_row = sheet.max_row + 1\n",
    "\n",
    "    # Append the sample data to the next empty row\n",
    "    for col, item in enumerate(sample_data, start=1):\n",
    "        sheet.cell(row=next_row, column=col).value = item\n",
    "\n",
    "    # Store test details in separate rows\n",
    "    tnum = len(test_data)\n",
    "    test_start_row = next_row\n",
    "\n",
    "    for i in range(tnum):\n",
    "        # Write the test data to the Excel workbook\n",
    "        row_values = test_data[i]\n",
    "        for col, item in enumerate(row_values, start=19):  # Starting from column 'S'\n",
    "            sheet.cell(row=next_row, column=col).value = item\n",
    "        next_row += 1\n",
    "\n",
    "    # Fill the remaining columns (A to S) with copies of the sample_data for each row of test_data\n",
    "    for col, item in enumerate(sample_data, start=1):\n",
    "        for row in range(test_start_row + 1, test_start_row + tnum):\n",
    "            sheet.cell(row=row, column=col).value = item\n",
    "    \n",
    "    # Save the changes back to the Excel file\n",
    "    wb.save(file_name)\n",
    "    \n",
    "    print('Your data are saved in the Excel Workbook.')\n",
    "\n",
    "    \n",
    "    \n",
    "def plot_time_sweep_data(df_ts):\n",
    "    \"\"\"\n",
    "    Plot the time sweep data as a scatter plot.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame containing the time sweep data.\n",
    "    \n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"\n",
    "    # Create the scatter plot\n",
    "    @interact(ymin=FloatRangeSlider(value=[1e-4, 1e2], min=1e-4, max=1e2, step=1e-4,\n",
    "                                    description='y-axis limits:', readout_format='.1e'),)\n",
    "    def plot_interactive(ymin):\n",
    "        sns.set(style='ticks')\n",
    "        custom_palette = sns.color_palette(\"deep\")\n",
    "        custom_palette[1] = \"red\"\n",
    "        sns.set_palette(custom_palette)\n",
    "        ax = sns.scatterplot(x='time (min)', y='G\\' (Pa)', data=df_ts, alpha=0.7)\n",
    "        sns.scatterplot(x='time (min)', y='G\\'\\' (Pa)', data=df_ts, alpha=0.7)\n",
    "        ax.set_ylim(ymin=ymin[0], ymax=ymin[1])\n",
    "        ax.set_yscale('log')\n",
    "        plt.grid(True, which=\"both\", ls=\"-\", c='white')\n",
    "        plt.xlabel(\"time (min)\")\n",
    "        plt.ylabel(\"G', G'' (Pa)\")\n",
    "        legend_artists = [plt.Line2D([0], [0], marker='o', lw=0),\n",
    "                          plt.Line2D([0], [0], marker='o', lw=0, color='red')]\n",
    "        legend_labels = ['G\\'', 'G\\'\\'']\n",
    "        ax.legend(legend_artists, legend_labels, bbox_to_anchor=(1.1, 1), loc='upper left')\n",
    "        plt.savefig('Results/time_sweep.png', dpi=300, bbox_inches='tight')\n",
    "        plt.show()\n",
    "\n",
    "        \n",
    "def time_sweep_data_analysis(df):\n",
    "    \"\"\"\n",
    "    Perform analysis on time sweep data for Osteoarthritis medical device.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame containing the time sweep data.\n",
    "\n",
    "    Returns:\n",
    "        mean_values (numpy.array): G' and G'' (Pa) at 3.142 rad/s from time sweep test.\n",
    "        std_values (numpy.array): G' and G'' standard deviations from time sweep test.\n",
    "    \"\"\"    \n",
    "    # Select the specific columns for the time sweep\n",
    "    columns_ts = ['time (min)', 'G\\' (Pa)', 'G\\'\\' (Pa)']\n",
    "    df_ts = df[columns_ts]\n",
    "    \n",
    "    # Plot\n",
    "    plot_time_sweep_data(df_ts)\n",
    "\n",
    "    # Determine the last time value\n",
    "    last_time_value = df_ts['time (min)'].iloc[-1]\n",
    "\n",
    "    # Prompt the user for start and finish time points\n",
    "    t_start = float(input(f'Give a time start point (0.1, 0.2 ... {last_time_value}): '))\n",
    "    t_finish = float(input(f'Give a time finish point (0.1, 0.2 ... {last_time_value}): '))\n",
    "\n",
    "    # Define the list of values within the selected time interval\n",
    "    time_interval = df_ts[(df_ts['time (min)'] >= t_start) & (df_ts['time (min)'] <= t_finish)]\n",
    "\n",
    "    # Calculate the mean and standard deviation of the selected values\n",
    "    mean_values = np.mean(time_interval, axis=0)[-2:]\n",
    "    std_values = np.std(time_interval, axis=0)[-2:]\n",
    "\n",
    "    # Create a table with two columns\n",
    "    table = PrettyTable(['Variable', 'Value'])\n",
    "\n",
    "    # Add the mean and standard deviation values to the table\n",
    "    table.add_row(['Mean', mean_values.to_string()])\n",
    "    table.add_row(['Standard deviation', std_values.to_string()])\n",
    "\n",
    "    # Print the table\n",
    "    print(table)\n",
    "\n",
    "    # Return the mean and standard deviation values as a DataFrame\n",
    "    result_df = pd.DataFrame({'Variable': ['Mean', 'Standard deviation'],\n",
    "                              'Value': [mean_values, std_values]})\n",
    "\n",
    "    return mean_values, std_values\n",
    "\n",
    "\n",
    "def plot_frequency_sweep_data(df_fs, frequency_interval):\n",
    "    \"\"\"\n",
    "    Plot the frequency sweep data as a scatter plot.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame containing the frequency sweep data.\n",
    "        frequency_interval (DataFrame): A subset of the df_fs DataFrame, representing the data within a selected time interval.\n",
    "    \n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"   \n",
    "    # Create the scatter plot\n",
    "    @interact(ymin_primary=FloatRangeSlider(value=[1e-4, 1e2], min=1e-4, max=1e2, step=1e-4, \n",
    "                                            description='Primary y-axis limits:', readout_format='.1e'),\n",
    "              ymin_secondary=FloatRangeSlider(value=[1e-4, 1e2], min=1e-4, max=1e2, step=1e-4, \n",
    "                                              description='Secondary y-axis limits:', readout_format='.1e'))\n",
    "    def show_graph(ymin_primary, ymin_secondary):\n",
    "        sns.set(style='ticks')\n",
    "        custom_palette = sns.color_palette(\"deep\")\n",
    "        custom_palette[1] = \"red\"\n",
    "        sns.set_palette(custom_palette)\n",
    "        ax = sns.scatterplot(x='ang. Frequency (rad/s)',y='G\\' (Pa).1', data=frequency_interval, alpha=0.7)\n",
    "        sns.scatterplot(x='ang. Frequency (rad/s)',y='G\\'\\' (Pa).1', data=frequency_interval, alpha=0.7)\n",
    "        ax.set_ylim(ymin=ymin_primary[0], ymax=ymin_primary[1])\n",
    "        ax.set_yscale('log') \n",
    "        ax.set_xscale('log')\n",
    "        plt.xlabel(\"angular frequency (rad/s)\")\n",
    "        plt.ylabel(\"G', G'' (Pa)\")\n",
    "        ax2 = ax.twinx()\n",
    "        sns.scatterplot(x='ang. Frequency (rad/s)',y='|n*| (Pa.s)', data=frequency_interval, color='green', \n",
    "                        ax=ax2, alpha=0.7)\n",
    "        ax2.set_ylim(ymin=ymin_secondary[0], ymax=ymin_secondary[1])\n",
    "        ax2.set_yscale('log')\n",
    "        ax.minorticks_on()\n",
    "        ax.grid(True, which='both')\n",
    "        plt.ylabel(\"|n*| (Pa.s)\")\n",
    "        legend_artists = [plt.Line2D([0], [0], marker='o', lw=0),\n",
    "                          plt.Line2D([0], [0], marker='o', color='red', lw=0),\n",
    "                          plt.Line2D([0], [0], marker='o', color='green', lw=0)]\n",
    "        legend_labels = ['G\\'', 'G\\'\\'', '|n*|']\n",
    "        ax.legend(legend_artists, legend_labels,bbox_to_anchor=(1.1, 1), loc='upper left')\n",
    "        plt.savefig('Results/frequency_sweep.png', dpi=300, bbox_inches='tight')\n",
    "        plt.show()\n",
    "\n",
    "    \n",
    "def frequency_sweep_data_analysis(df):\n",
    "    \"\"\"\n",
    "    Perform analysis on frequency sweep data for Osteoarthritis medical device.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame containing the frequency sweep data.\n",
    "\n",
    "    Returns:\n",
    "        G_pf (float): G' (Pa) at 3.142 rad/s from frequency sweep test.\n",
    "        G_dpf (float): G'' (Pa) at 3.142 rad/s from frequency sweep test.\n",
    "        G_pf1 (float): G' (Pa) at 0.68 rad/s from frequency sweep test.\n",
    "        G_dpf1 (float): G'' (Pa) at 0.68 rad/s from frequency sweep test.\n",
    "        G_pf2 (float): G' (Pa) at 14.58 rad/s from frequency sweep test.\n",
    "        G_dpf2 (float): G'' (Pa) at 14.58 rad/s from frequency sweep test.\n",
    "        eta_star (float): Complex viscosity (Pa.s) at 3.142 rad/s from frequency sweep test.\n",
    "        crossover_point (float): Cross-over point from frequency sweep test.\n",
    "    \"\"\"\n",
    "    # Select the specific columns for the frequency sweep\n",
    "    columns_fs = ['ang. Frequency (rad/s)', 'G\\' (Pa).1', 'G\\'\\' (Pa).1', '|n*| (Pa.s)', 'delta (degrees)']\n",
    "    df_fs = df[columns_fs]\n",
    "    \n",
    "    # Delete NaN values\n",
    "    df_fs = df_fs.dropna(how='any')\n",
    "    \n",
    "    # Define the list of values within the selected delta range\n",
    "    frequency_interval = df_fs[df_fs['delta (degrees)'] < 90]\n",
    "    delta_df = df_fs[df_fs['delta (degrees)'] < 150]\n",
    "    \n",
    "    # Plot\n",
    "    plot_frequency_sweep_data(df_fs, frequency_interval)\n",
    "    \n",
    "    # Get the frequency sweep data at specific indices\n",
    "    G_pf = frequency_interval.loc[3, 'G\\' (Pa).1']\n",
    "    G_dpf = frequency_interval.loc[3, 'G\\'\\' (Pa).1']\n",
    "    G_pf1 = frequency_interval.loc[1, 'G\\' (Pa).1']\n",
    "    G_dpf1 = frequency_interval.loc[1, 'G\\'\\' (Pa).1']\n",
    "    \n",
    "    # Get the frequency sweep data at specific indices if delta <= 90\n",
    "    delta = delta_df.loc[5, 'delta (degrees)']\n",
    "    if delta <= 90:\n",
    "        G_pf2 = frequency_interval.loc[5, 'G\\' (Pa).1']\n",
    "        G_dpf2 = frequency_interval.loc[5, 'G\\'\\' (Pa).1']\n",
    "    else:\n",
    "        G_pf2 = np.nan\n",
    "        G_dpf2 = np.nan\n",
    "    \n",
    "    # Get the viscosity |n*| at 3.142 rad/s\n",
    "    eta_star = frequency_interval.loc[3, '|n*| (Pa.s)']\n",
    "    \n",
    "    # Check which value is bigger in the first row (G' or G'')\n",
    "    if frequency_interval.iloc[0]['G\\' (Pa).1'] > frequency_interval.iloc[0]['G\\'\\' (Pa).1']:\n",
    "        bigger_col = 'G\\' (Pa).1'\n",
    "        smaller_col = 'G\\'\\' (Pa).1'\n",
    "    else:\n",
    "        bigger_col = 'G\\'\\' (Pa).1'\n",
    "        smaller_col = 'G\\' (Pa).1'\n",
    "\n",
    "    # Initialize variables to store the closest difference\n",
    "    closest_diff = float('inf')\n",
    "    closest_diff_index = None\n",
    "\n",
    "    # Find the crossover point\n",
    "    for index, row in frequency_interval.iterrows():\n",
    "        # Check if the bigger value crossed the smaller value\n",
    "        if row[bigger_col] < row[smaller_col]:\n",
    "            # Calculate the difference between the values\n",
    "            diff = abs(row[bigger_col] - row[smaller_col])\n",
    "            # Check if this difference is closer than the previous closest difference\n",
    "            if diff < closest_diff:\n",
    "                closest_diff = diff\n",
    "                closest_diff_index = index\n",
    "\n",
    "    # Check if a crossover point was found\n",
    "    if closest_diff_index is not None:\n",
    "        crossover_point = frequency_interval.loc[closest_diff_index, 'ang. Frequency (rad/s)']\n",
    "        print(f\"Crossover point at: {crossover_point}\")\n",
    "    else:\n",
    "        crossover_point = None\n",
    "        print(\"No crossover point was found.\")\n",
    "    \n",
    "    print()\n",
    "    return G_pf, G_dpf, G_pf1, G_dpf1, G_pf2, G_dpf2, eta_star, crossover_point\n",
    "\n",
    "\n",
    "def calculate_cross_model_equation(s, eta_0, eta_inf, K, n):\n",
    "    \"\"\"\n",
    "    Calculate the cross model equation given the parameters.\n",
    "\n",
    "    Parameters:\n",
    "        s (numpy.array): The array of shear rates.\n",
    "        eta_0 (float): The zero-rate viscosity (Pa.s).\n",
    "        eta_inf (float): The infinite-rate viscosity (Pa.s).\n",
    "        K (float): The consistency (s).\n",
    "        n (float): The rate index (-).\n",
    "\n",
    "    Returns:\n",
    "        numpy.array: The calculated viscosities using the cross model equation.\n",
    "    \"\"\"\n",
    "    return (eta_0 - eta_inf) / (1 + (s * K) ** n) + eta_inf\n",
    "\n",
    "\n",
    "def plot_flow_step_data(df_step, s_cross, n_cross, ymin_primary):\n",
    "    \"\"\"\n",
    "    Plot the flow step data as a scatter plot.\n",
    "\n",
    "    Parameters:\n",
    "        df_step (DataFrame): The input DataFrame containing the flow step data.\n",
    "        s_cross (numpy.array): The array of shear rates for cross model calculation.\n",
    "        n_cross (numpy.array): The calculated viscosities using the cross model equation.\n",
    "        ymin_primary (tuple): The range for the primary y-axis.\n",
    "    \"\"\"\n",
    "    sns.set(style='ticks')\n",
    "    custom_palette = sns.color_palette(\"deep\")\n",
    "    custom_palette[1] = \"red\"\n",
    "    sns.set_palette(custom_palette)\n",
    "    ax = sns.lineplot(x=s_cross, y=n_cross, alpha=0.7)\n",
    "    sns.scatterplot(x='shear rate (1/s)', y='viscosity (Pa.s)', data=df_step, color=\"red\", alpha=0.7)\n",
    "    ax.set_xlim(xmin=1e-4, xmax=1e6)\n",
    "    ax.set_ylim(ymin=ymin_primary[0], ymax=ymin_primary[1])\n",
    "    ax.set_yscale('log')\n",
    "    ax.set_xscale('log')\n",
    "    ax.grid(True, which='both')\n",
    "    plt.xlabel(\"shear rate (1/s)\")\n",
    "    plt.ylabel(\"viscosity (Pa.s)\")\n",
    "    legend_artists = [plt.Line2D([0], [0], marker='o', lw=0),\n",
    "                      plt.Line2D([0], [0], color='red', lw=4)]\n",
    "    legend_labels = ['Cross model viscosity', 'Flow step viscosity']\n",
    "    ax.legend(legend_artists, legend_labels, bbox_to_anchor=(1.1, 1), loc='upper left')\n",
    "    plt.savefig('Results/flow_step.png', dpi=300, bbox_inches='tight')\n",
    "    plt.show()\n",
    "\n",
    "    \n",
    "def flow_step_data_analysis(df):\n",
    "    \"\"\"\n",
    "    Perform analysis on flow step data for Osteoarthritis medical device.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame containing the flow step data.\n",
    "\n",
    "    Returns:\n",
    "        eta_0 (float): Zero rate viscosity (Pa.s) from cross model.\n",
    "        eta_inf (float): Infinite rate viscosity (Pa.s) from cross model.\n",
    "        K (float): Consistency (Pa.s) from cross model.\n",
    "        n (float): Exponent (-) from cross model.\n",
    "        error (float): Deviation of the model (-) from cross model.\n",
    "        eta (float): Viscosity (Pa.s) at 3.162 1/s from flow step measurement.\n",
    "    \"\"\"\n",
    "    # Select the specific columns for the flow step\n",
    "    columns_step = ['shear rate (1/s)', 'viscosity (Pa.s)']\n",
    "    df_step = df[columns_step]\n",
    "    \n",
    "    # Delete NaN values\n",
    "    df_step = df_step.dropna(how='any')\n",
    "\n",
    "    # Extract the data for cross model calculation\n",
    "    s = df_step['shear rate (1/s)']\n",
    "    eta_0 = df.at[0, 'zero-rate viscosity (Pa.s)']\n",
    "    eta_inf = df.at[0, 'infinite-rate viscosity (Pa.s)']\n",
    "    K = df.at[0, 'consistency (s)']\n",
    "    n = df.at[0, 'rate index (-)']\n",
    "    eta = df_step.at[3, 'viscosity (Pa.s)']\n",
    "\n",
    "    # Calculate the cross model equation\n",
    "    s_start = np.array([0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05])\n",
    "    s_end = np.array([5000, 10000, 20000, 50000, 100000, 200000, 500000, 1000000])\n",
    "    s_cross = np.concatenate((s_start, s, s_end))\n",
    "    n_cross = calculate_cross_model_equation(s_cross, eta_0, eta_inf, K, n)\n",
    "\n",
    "    # Create the interactive plot\n",
    "    @interact(ymin_primary=FloatRangeSlider(value=[1e-4, 1e2], min=1e-4, max=1e2, step=1e-4,\n",
    "                                            description='y-axis limits:', readout_format='.1e'),)\n",
    "    def plot_interactive(ymin_primary):\n",
    "        plot_flow_step_data(df_step, s_cross, n_cross, ymin_primary)\n",
    "\n",
    "    # Return the calculated values as a dictionary\n",
    "    return eta_0, eta_inf, K, n, df.at[0, 'standard error'], eta\n",
    "\n",
    "\n",
    "def plot_cox_merz_data(df_cox, ymin_primary):\n",
    "    \"\"\"\n",
    "    Plot the Cox-Merz data as a scatter plot.\n",
    "\n",
    "    Parameters:\n",
    "        df_cox (DataFrame): The input DataFrame containing the Cox-Merz data.\n",
    "        ymin_primary (tuple): The range for the primary y-axis.\n",
    "    \n",
    "    Returns:\n",
    "        None\n",
    "    \"\"\"\n",
    "    sns.set(style='ticks')\n",
    "    custom_palette = sns.color_palette(\"deep\")\n",
    "    custom_palette[1] = \"green\"\n",
    "    sns.set_palette(custom_palette)\n",
    "    ax = sns.scatterplot(x='shear rate (1/s)', y='viscosity (Pa.s)', data=df_cox, alpha=0.7)\n",
    "    sns.scatterplot(x='ang. Frequency (rad/s)', y='|n*| (Pa.s)', data=df_cox, alpha=0.7)\n",
    "    ax.set_ylim(ymin=ymin_primary[0], ymax=ymin_primary[1])\n",
    "    ax.set_yscale('log')\n",
    "    ax.set_xscale('log')\n",
    "    ax.grid(True, which='both')\n",
    "    plt.xlabel(\"shear rate (1/s), angular frequency (rad/s)\")\n",
    "    plt.ylabel(\"viscosity (Pa.s), |n*| (Pa.s)\")\n",
    "    legend_artists = [plt.Line2D([0], [0], marker='o', lw=0),\n",
    "                      plt.Line2D([0], [0], marker='o', color='green', lw=0)]\n",
    "    legend_labels = ['viscosity', '|n*|']\n",
    "    ax.legend(legend_artists, legend_labels, bbox_to_anchor=(1.1, 1), loc='upper left')\n",
    "    plt.savefig('Results/cox_merz.png', dpi=300, bbox_inches='tight')\n",
    "    plt.show()\n",
    "\n",
    "    \n",
    "def cox_merz_rule(df):\n",
    "    \"\"\"\n",
    "    Perform Cox-Merz analysis for Osteoarthritis medical device.\n",
    "\n",
    "    Parameters:\n",
    "        df_cox (DataFrame): The input DataFrame containing the Cox-Merz data.\n",
    "\n",
    "    Returns:\n",
    "        eta (float): Viscosity (Pa.s) at 3.162 1/s from flow step test.\n",
    "        eta_star (float): Complex viscosity (Pa.s) at 3.142 rad/s from frequency test.\n",
    "    \"\"\"\n",
    "    # Select the specific columns for the cox merz\n",
    "    columns_cox = ['shear rate (1/s)', 'viscosity (Pa.s)', 'ang. Frequency (rad/s)', '|n*| (Pa.s)']\n",
    "    df_cox = df[columns_cox]\n",
    "    # Delete NaN values\n",
    "    df_cox = df_cox.dropna(how='any')\n",
    "\n",
    "    # Create the interactive plot\n",
    "    @interact(ymin_primary=FloatRangeSlider(value=[1e-4, 1e2], min=1e-4, max=1e2, step=1e-4,\n",
    "                                            description='y-axis limits:', readout_format='.1e'),)\n",
    "    def plot_interactive(ymin_primary):\n",
    "        plot_cox_merz_data(df_cox, ymin_primary)\n",
    "\n",
    "    # Get the viscosities at specific shear rate and angular frequency\n",
    "    eta = df_cox.at[3, 'viscosity (Pa.s)']\n",
    "    eta_star = df_cox.at[3, '|n*| (Pa.s)']\n",
    "\n",
    "    # Create a table with three columns\n",
    "    table = PrettyTable(['Experiment', 'Variable', 'Value'])\n",
    "\n",
    "    # Add the mean and standard deviation values to the table\n",
    "    table.add_row(['Flow step', 'viscosity (Pa.s)', eta])\n",
    "    table.add_row(['Frequency sweep', '|n*| (Pa.s)', eta_star])\n",
    "\n",
    "    # Print the table\n",
    "    print(table)\n",
    "\n",
    "    # Return the calculated values as a dictionary\n",
    "    return eta, eta_star\n",
    "\n",
    "\n",
    "def sample_analysis(sid, df_ed=\"Data/Experimental_Data.xlsx\"):\n",
    "    \"\"\"\n",
    "    Perform sample analysis and generate a report.\n",
    "\n",
    "    Parameters:\n",
    "        sid (str): The sample ID.\n",
    "        df_ed (DataFrame, optional): The input DataFrame containing the experimental data. Default value: \"Data/Experimental_Data.xlsx\".\n",
    "\n",
    "    Returns:\n",
    "        dict: A dictionary containing the analysis results.\n",
    "    \"\"\"    \n",
    "    df_ed = pd.read_excel(df_ed, sid)\n",
    "    analysis_results = {}\n",
    "    \n",
    "    # Check if the time sweep data is present in the \"df_ed\" dataframe\n",
    "    if \"time (min)\" in df_ed.columns:\n",
    "        mean_values, std_values = time_sweep_data_analysis(df_ed)\n",
    "        analysis_results['time_sweep'] = {\n",
    "            'mean_values': mean_values,\n",
    "            'std_values': std_values\n",
    "        }\n",
    "    else:\n",
    "        print(\"There are no time sweep data\")\n",
    "\n",
    "    # Check if the frequency sweep data is present in the \"df_ed\" dataframe\n",
    "    if \"ang. Frequency (rad/s)\" in df_ed.columns:\n",
    "        G_pf, G_dpf, G_pf1, G_dpf1, G_pf2, G_dpf2, eta_star, crossover_point = frequency_sweep_data_analysis(df_ed)\n",
    "        analysis_results['frequency_sweep'] = {\n",
    "            'G_pf': G_pf,\n",
    "            'G_dpf': G_dpf,\n",
    "            'G_pf1': G_pf1,\n",
    "            'G_dpf1': G_dpf1,\n",
    "            'G_pf2': G_pf2,\n",
    "            'G_dpf2': G_dpf2,\n",
    "            'eta_star': eta_star,\n",
    "            'crossover_point': crossover_point\n",
    "        }\n",
    "    else:\n",
    "        print(\"There are no frequency sweep data\")\n",
    "\n",
    "    # Check if the flow step data is present in the \"df_ed\" dataframe\n",
    "    if \"shear rate (1/s)\" in df_ed.columns:\n",
    "        eta_0, eta_inf, K, n, error, eta = flow_step_data_analysis(df_ed)\n",
    "        analysis_results['flow_step'] = {\n",
    "        'eta_0': eta_0,\n",
    "        'eta_inf': eta_inf,\n",
    "        'K': K,\n",
    "        'n': n,\n",
    "        'error': error,\n",
    "        'eta': eta\n",
    "        }\n",
    "    else:\n",
    "        print(\"There are no flow step data\")\n",
    "\n",
    "    # Check if the cox merz data is present in the \"df_ed\" dataframe\n",
    "    if \"ang. Frequency (rad/s)\" in df_ed.columns and \"shear rate (1/s)\" in df_ed.columns:\n",
    "        eta, eta_star = cox_merz_rule(df_ed)\n",
    "        analysis_results['cox_merz'] = {\n",
    "        'eta': eta,\n",
    "        'eta_star': eta_star\n",
    "        }\n",
    "    else:\n",
    "        print(\"There are no cox merz data\")\n",
    "        \n",
    "    return analysis_results\n",
    "\n",
    "\n",
    "\n",
    "def append_samples_to_final_database(start_row_n, database_file=\"Data/Database.xlsx\", experiment_data_file=\"Data/Experimental_Data.xlsx\", analysis_data_file=\"Data/Statistical_Analysis_Database.xlsx\"):\n",
    "    \"\"\"\n",
    "    Load the database file, process the data, and store the analysis results.\n",
    "\n",
    "    Parameters:\n",
    "        start_row_n (int): The sample number from which to start the analysis.\n",
    "        database_file (str, optional): The filename of the database Excel file. Default value: \"Data/Database.xlsx\".\n",
    "        experiment_data_file (str, optional): The folder path containing the experiment data files. Default value: \"Data/Experimental_Data.xlsx\".\n",
    "        analysis_data_file (str, optional): Name of the Excel file to be updated. Default value: \"Data/Statistical_Analysis_Database.xlsx\".\n",
    "\n",
    "    Returns:\n",
    "        DataFrame: The updated database with the analysis results.\n",
    "    \"\"\"\n",
    "    start_row_n = str(start_row_n)\n",
    "    \n",
    "    # Load the database file into a DataFrame\n",
    "    database = pd.read_excel(database_file)\n",
    "\n",
    "    # Set the \"ID\" column as the index of the dataframe\n",
    "    database = database.set_index(\"ID\")\n",
    "\n",
    "    # Remove any duplicates in the \"ID\" index, keeping the first occurrence\n",
    "    database = database[~database.index.duplicated(keep='first')]\n",
    "\n",
    "    # Remove the columns you don't need\n",
    "    to_drop = [\"Date\", \"Total Volume (mL)\", \"Color\", \"Transparency\", \"Texture\", \"Number of Tests\", \"Test Type\", \"Temperature (oC)\", \"Bubbles before Test\", \"Bubbles after Test\"]\n",
    "    database = database.drop(columns=to_drop)\n",
    "\n",
    "    # Show the resulting dataframe\n",
    "    print(database)\n",
    "\n",
    "    # Start row\n",
    "    start_row = 'S' + start_row_n\n",
    "    print(f\"The appending starts from the {start_row} sample.\")\n",
    "\n",
    "    # Slice the DataFrame to select only the rows after the start row\n",
    "    sub_database = database.loc[start_row:].copy()\n",
    "    \n",
    "    # Loop through each ID (row) in the dataframe\n",
    "    for index, row in sub_database.iterrows():\n",
    "        sid = index  # Set ID as sid\n",
    "        print(f'Sample ID: {sid}')\n",
    "        \n",
    "        analysis_results = sample_analysis(sid, experiment_data_file)\n",
    "\n",
    "        if \"time_sweep_mean\" in analysis_results:\n",
    "            time_results = analysis_results[\"time_sweep\"]\n",
    "            sub_database.loc[sid, \"Time Sweep G' average (Pa)\"] = time_results[\"time_sweep_mean\"][0]\n",
    "            sub_database.loc[sid, \"Time Sweep G'' average (Pa)\"] = time_results[\"time_sweep_mean\"][1]\n",
    "            sub_database.loc[sid, \"Time Sweep G' standard deviation (-)\"] = time_results[\"time_sweep_std\"][0]\n",
    "            sub_database.loc[sid, \"Time Sweep G'' standard deviation (-)\"] = time_results[\"time_sweep_std\"][1]\n",
    "\n",
    "        if \"frequency_sweep\" in analysis_results:\n",
    "            freq_results = analysis_results[\"frequency_sweep\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G' at 0.68 rad/s (Pa)\"] = freq_results[\"G_pf1\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G'' at 0.68 rad/s (Pa)\"] = freq_results[\"G_dpf1\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G' at 3.142 rad/s (Pa)\"] = freq_results[\"G_pf\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G'' at 3.142 rad/s (Pa)\"] = freq_results[\"G_dpf\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G' at 14.58 rad/s (Pa)\"] = freq_results[\"G_pf2\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep G'' at 14.58 rad/s (Pa)\"] = freq_results[\"G_dpf2\"]\n",
    "            sub_database.loc[sid, \"Cross Over Point (rad/s)\"] = freq_results[\"crossover_point\"]\n",
    "            sub_database.loc[sid, \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\"] = freq_results[\"eta_star\"]\n",
    "\n",
    "        if \"flow_step\" in analysis_results:\n",
    "            flow_results = analysis_results[\"flow_step\"]\n",
    "            sub_database.loc[sid, \"Zero-rate Viscosity (Pa.s)\"] = flow_results[\"eta_0\"]\n",
    "            sub_database.loc[sid, \"Infinite-rate Viscosity (Pa.s)\"] = flow_results[\"eta_inf\"]\n",
    "            sub_database.loc[sid, \"Consistency (s)\"] = flow_results[\"K\"]\n",
    "            sub_database.loc[sid, \"Rate Index (-)\"] = flow_results[\"n\"]\n",
    "            sub_database.loc[sid, \"Standard Error (-)\"] = flow_results[\"error\"]\n",
    "            sub_database.loc[sid, \"Flow Step Viscosity at 3.162 1/s (Pa.s)\"] = flow_results[\"eta\"]\n",
    "\n",
    "        print(sub_database.loc[sid])\n",
    "\n",
    "    print(sub_database)\n",
    "    \n",
    "    # Reset the index to move the \"ID\" column back as a regular column\n",
    "    sub_database.reset_index(inplace=True)\n",
    "    \n",
    "    existing_database = pd.read_excel(analysis_data_file)\n",
    "    result = pd.concat([existing_database, sub_database], ignore_index=True)\n",
    "    result = result.reindex()\n",
    "\n",
    "    result.to_excel(analysis_data_file, index=False)\n",
    "\n",
    "    return result\n",
    "\n",
    "\n",
    "\n",
    "def create_sample_report(snum, \n",
    "                         database_file=\"Data/Database.xlsx\", \n",
    "                         analysis_data_file=\"Data/Statistical_Analysis_Database.xlsx\", \n",
    "                         txt_file=\"Data/report_1/report_template_1.txt\", \n",
    "                         t_start=0.5, \n",
    "                         t_finish=5\n",
    "                        ):\n",
    "    \"\"\"\n",
    "    Load and process the sample's data. Create the technical report with the results.\n",
    "\n",
    "    Parameters:\n",
    "        snum (int): The sample ID number.\n",
    "        database_file (str, optional): The filename of the database Excel file. Default value: \"Data/Database.xlsx\".\n",
    "        analysis_data_file (str, optional): The filename of the Excel file containing the statistical analysis data. The default value is \"Data/Statistical_Analysis_Database.xlsx\".\n",
    "        txt_file (str, optional): The filename of the text file serving as a report template. Default value: \"report_template.txt\".\n",
    "        t_start (float, optional): The starting time in minutes for data analysis in the Time Sweep test. Default value: 0.5.\n",
    "        t_finish (float, optional): The ending time in minutes for data analysis in the Time Sweep test. Default value: 5.\n",
    "    \n",
    "    Returns:\n",
    "        None.\n",
    "    \"\"\"\n",
    "    sid = 'S' + str(snum)\n",
    "    \n",
    "    # Load the database files into DataFrames\n",
    "    database = pd.read_excel(database_file)\n",
    "    database = database.set_index(\"ID\")\n",
    "    database = database[~database.index.duplicated(keep='first')]\n",
    "    statistical_database = pd.read_excel(analysis_data_file)\n",
    "    statistical_database = statistical_database.set_index(\"ID\")\n",
    "\n",
    "    # Data from database_file\n",
    "    selected_sample = database.loc[sid]\n",
    "    sdate = selected_sample['Date']\n",
    "    sgender = selected_sample['Gender']\n",
    "    sage = selected_sample['Age']\n",
    "    sjoint = selected_sample['Joint']\n",
    "    scon = selected_sample['Condition']\n",
    "    svol = selected_sample['Total Volume (mL)']\n",
    "    scolor = selected_sample['Color']\n",
    "    strans = selected_sample['Transparency']\n",
    "    stexture = selected_sample['Texture']\n",
    "    sblood = selected_sample['Blood']\n",
    "    sclot = selected_sample['Clot']\n",
    "    stissue = selected_sample['Tissue']\n",
    "    frstorage = selected_sample['Storage in freezer']\n",
    "    covid = selected_sample['COVID 19']\n",
    "    vaccine = selected_sample['Vaccined COVID 19']\n",
    "    covid_now = selected_sample['COVID 19 now']\n",
    "    \n",
    "    # Data from analysis_data_file\n",
    "    selected_sample2 = statistical_database.loc[sid]\n",
    "    G_pt = selected_sample2[\"Time Sweep G' average (Pa)\"]\n",
    "    G_dpt = selected_sample2[\"Time Sweep G'' average (Pa)\"]\n",
    "    G_pstd = selected_sample2[\"Time Sweep G' standard deviation (-)\"]\n",
    "    G_dpstd = selected_sample2[\"Time Sweep G'' standard deviation (-)\"]\n",
    "    G_pf1 = selected_sample2[\"Frequency Sweep G' at 0.68 rad/s (Pa)\"]\n",
    "    G_dpf1 = selected_sample2[\"Frequency Sweep G'' at 0.68 rad/s (Pa)\"]\n",
    "    G_pf = selected_sample2[\"Frequency Sweep G' at 3.142 rad/s (Pa)\"]\n",
    "    G_dpf = selected_sample2[\"Frequency Sweep G'' at 3.142 rad/s (Pa)\"]\n",
    "    G_pf2 = selected_sample2[\"Frequency Sweep G' at 14.58 rad/s (Pa)\"]\n",
    "    G_dpf2 = selected_sample2[\"Frequency Sweep G'' at 14.58 rad/s (Pa)\"]\n",
    "    crossover_point = selected_sample2[\"Cross Over Point (rad/s)\"]\n",
    "    eta_star = selected_sample2[\"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\"]\n",
    "    eta_0 = selected_sample2[\"Zero-rate Viscosity (Pa.s)\"]\n",
    "    eta_inf = selected_sample2[\"Infinite-rate Viscosity (Pa.s)\"]\n",
    "    K = selected_sample2[\"Consistency (s)\"]\n",
    "    n = selected_sample2[\"Rate Index (-)\"]\n",
    "    error = selected_sample2[\"Standard Error (-)\"]\n",
    "    eta = selected_sample2[\"Flow Step Viscosity at 3.162 1/s (Pa.s)\"]\n",
    "\n",
    "    # Data from report_template.txt\n",
    "    variables = {}\n",
    "    with open(txt_file, \"r\") as file:\n",
    "        for line in file:\n",
    "            # Use regular expression to find lines with variables and values\n",
    "            match = re.match(r\"^\\s*([\\w_]+)\\s*=\\s*(.+)\\s*$\", line)\n",
    "            if match:\n",
    "                variable_name, value = match.groups()\n",
    "                variables[variable_name] = value\n",
    "    \n",
    "    # Format information\n",
    "    font = variables[list(variables.keys())[0]]\n",
    "\n",
    "    # Basic information\n",
    "    writer_name = variables[list(variables.keys())[1]]\n",
    "    sup_name = variables[list(variables.keys())[2]]\n",
    "    today = datetime.date.today()\n",
    "    today_date = today.strftime(\"%d/%m/%Y\")\n",
    "    today_date_str = today_date.replace(\"/\", \"_\")\n",
    "    filename = f\"{sid}_Report_{today_date_str}.pdf\"\n",
    "    header = f\"HSF_{sid}_Report\"\n",
    "    title = f\"Report - HSF_{sid}\"\n",
    "    \n",
    "    # Images\n",
    "    im1 = variables[list(variables.keys())[3]]\n",
    "    im2 = variables[list(variables.keys())[4]]\n",
    "    im3 = variables[list(variables.keys())[5]]\n",
    "    im4 = variables[list(variables.keys())[6]]\n",
    "    im5 = variables[list(variables.keys())[7]]\n",
    "    im6 = variables[list(variables.keys())[8]]\n",
    "    im7 = variables[list(variables.keys())[9]]\n",
    "    im8 = variables[list(variables.keys())[10]]\n",
    "    im9 = variables[list(variables.keys())[11]]\n",
    "    im10 = variables[list(variables.keys())[12]]\n",
    "    im11 = variables[list(variables.keys())[13]]\n",
    "    im12 = variables[list(variables.keys())[14]]\n",
    "\n",
    "    # Other information\n",
    "    test_temp = variables[list(variables.keys())[15]]\n",
    "    \n",
    "    # Text\n",
    "    time_text = variables[list(variables.keys())[16]]\n",
    "    freq_text = variables[list(variables.keys())[17]]\n",
    "    flow_text = variables[list(variables.keys())[18]]\n",
    "    cox_text = variables[list(variables.keys())[19]]\n",
    "    results_text = variables[list(variables.keys())[20]]\n",
    "    \n",
    "    # Control Loops\n",
    "    if sblood == \"yes\":\n",
    "        if sclot == \"yes\":\n",
    "            if stissue == \"yes\":\n",
    "                bct = \"with blood, clot and tissue\"\n",
    "            else:\n",
    "                bct = \"with blood and clot, but without tissue\"\n",
    "        else:\n",
    "            if stissue == \"yes\":\n",
    "                bct = \"with blood and tissue, but without clot\"\n",
    "            else:\n",
    "                bct = \"with blood, but without tissue or clot\"\n",
    "    else:\n",
    "        if stissue == \"yes\":\n",
    "            bct = \"without blood, but with tissue\"\n",
    "        else:\n",
    "            bct = \"without blood, clot and tissue\"\n",
    "\n",
    "    # Header and footer\n",
    "    class PDF(FPDF):\n",
    "        def __init__(self):\n",
    "            super().__init__()\n",
    "        def header(self):\n",
    "            if self.page_no() != 1:\n",
    "                self.set_font(font, '', 10)\n",
    "                self.set_text_color(128, 128, 128) # set color to grey\n",
    "                self.cell(0, 10, f\"© Biomedical Engineering Lab | Chemical Engineering AUTh - {header}\", 0, 1, 'L')\n",
    "        def footer(self):\n",
    "            self.set_y(-15)\n",
    "            self.set_font(font, '', 10)\n",
    "            self.set_text_color(128, 128, 128) # set color to grey\n",
    "            self.cell(0, 10, f'Page {self.page_no()}', 0, 0, 'C')\n",
    "\n",
    "    # Report\n",
    "    pdf = PDF()\n",
    "    pdf.add_page()\n",
    "\n",
    "    # Title section\n",
    "    pdf.set_font(font, 'B', 16)\n",
    "    pdf.cell(w=0, h=10, txt=title, ln=1, align='C')\n",
    "    pdf.ln(5)\n",
    "\n",
    "    # Report's information\n",
    "    pdf.set_font(font, '', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"Author: {writer_name}\", ln=1)\n",
    "    pdf.set_font(font, '', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"Supervisor: {sup_name}\", ln=1)\n",
    "    pdf.set_font(font, '', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"Date: {today_date}\", ln=1)\n",
    "    pdf.ln(10)\n",
    "\n",
    "\n",
    "    # Sample's information\n",
    "    pdf.set_font(font, 'B', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"1. Sample's Information\", ln=1)\n",
    "    pdf.set_font(font, '', 12)\n",
    "    tnum = 1\n",
    "    pdf.set_font(font, 'I', 10)\n",
    "    pdf.multi_cell(w=0, h=10, txt=f\"Table {tnum}: Sample's information summary.\")\n",
    "    pdf.set_font(font, '', 12)\n",
    "    data = [['Sampling date', sdate],\n",
    "            ['Test date', today_date],\n",
    "            ['Stored in freezer', frstorage],\n",
    "            ['COVID 19 in general', covid],\n",
    "            ['COVID 19 vaccination', vaccine],\n",
    "            ['COVID 19 when tested', covid_now]]\n",
    "    # Set width and height of cells\n",
    "    cell_width = 60\n",
    "    cell_height = 6\n",
    "    # Loop through data and add cells to the table\n",
    "    for row in data:\n",
    "        for item in row:\n",
    "            pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "        pdf.ln()\n",
    "    pdf.ln(10)\n",
    "\n",
    "\n",
    "    # Sample's description\n",
    "    pdf.set_font(font, 'B', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"2. Sample's Description\", ln=1)\n",
    "    pdf.set_font(font, '', 12)\n",
    "    pdf.multi_cell(w=0, h=10, txt=f\"The sample had a total volume of approximately {svol}\" + \\\n",
    "                   f\" ml. It was {scolor} in color, {strans}, {stexture}, {bct}.\", border=0, align='J')\n",
    "\n",
    "    pdf.image(im1, x=pdf.get_x(), y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "    pdf.image(im2, x=pdf.get_x() + 100, y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "    pdf.ln(100)\n",
    "\n",
    "    fnum = 1\n",
    "    pdf.set_font(font, 'I', 10)\n",
    "    pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. HSF_{sid} in syringe.\", ln=2, align='C')\n",
    "    fnum = fnum+1\n",
    "\n",
    "    pdf.add_page()\n",
    "\n",
    "    # Analysis\n",
    "    pdf.set_font(font, 'B', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"3. Analysis\", ln=1)\n",
    "\n",
    "    if not math.isnan(G_pt):\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Time Sweep test at {test_temp} oC: \" + \\\n",
    "                       f\"The variation of G', G'' (Pa) over time (min) at {test_temp} oC with\" + \\\n",
    "                       \" an oscillation frequency of 3.142 rad/s and a 3% shear stress.\", border=0, align='J')\n",
    "        pdf.image(im9, x=pdf.get_x()+19, y=pdf.get_y()+5, w=0.8*pdf.w, h=0.5*pdf.w)\n",
    "        pdf.ln(110)\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. Time sweep test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"The means and standard deviation are calculated for the data from {t_start}\" + \\\n",
    "                       f\" to {t_finish} minutes.\", border=0, align='J')\n",
    "        tnum = tnum+1\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Table {tnum}: Time sweep analysis results.\")\n",
    "        pdf.set_font(font, '', 12)\n",
    "        data = [['Mean G\\'', round(G_pt,4)],\n",
    "                ['Standard Deviation G\\'', f\"{round(G_pstd,6)*100} %\"],\n",
    "                ['Mean G\\'\\'', round(G_dpt,4)],\n",
    "                ['Standard Deviation G\\'\\'', f\"{round(G_dpstd,6)*100} %\"]]\n",
    "        # Set width and height of cells\n",
    "        cell_width = 60\n",
    "        cell_height = 6\n",
    "        # Loop through data and add cells to the table\n",
    "        for row in data:\n",
    "            for item in row:\n",
    "                pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "            pdf.ln()\n",
    "        pdf.add_page()\n",
    "\n",
    "        pdf.image(im3, x=pdf.get_x(), y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.image(im4, x=pdf.get_x() + 100, y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.ln(100)\n",
    "\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. HSF_{sid} before (left) and after (right)\" + \\\n",
    "                 f\" Time Sweep test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "\n",
    "        pdf.set_font(font, 'U', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Author's notes:\", border=0, align='J')\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=time_text, border=0, align='J')\n",
    "        pdf.add_page()\n",
    "\n",
    "\n",
    "    if not math.isnan(G_pf):\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Frequency Sweep test at {test_temp} oC: \" + \\\n",
    "                       f\"The variation of G', G'' (Pa) and |n*| (Pa s) as a function of the \" + \\\n",
    "                       f\"oscillation frequency of the strain (rad/s) at {test_temp} oC.\", border=0, align='J')\n",
    "        pdf.image(im10, x=pdf.get_x()+19, y=pdf.get_y()+5, w=0.8*pdf.w, h=0.5*pdf.w)\n",
    "        pdf.ln(110)\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. Frequency sweep test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "        pdf.set_font(font, '', 12)\n",
    "        if not math.isnan(crossover_point):\n",
    "            pdf.multi_cell(w=0, h=10, txt=f\"Cross over point at {crossover_point} rad/s.\", border=0, align='J')\n",
    "        tnum = tnum+1\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Table {tnum}: Frequency sweep analysis results.\")\n",
    "        pdf.set_font(font, '', 12)\n",
    "        data = [['Time sweep G\\'', round(G_pt,4)],\n",
    "                ['Frequency sweep G\\'', round(G_pf,4)],\n",
    "                ['Percentage difference', f\"{round(abs(G_pt-G_pf)/((G_pt+G_pf)/2),6)*100} %\"],\n",
    "                ['Time sweep G\\'\\'', round(G_dpt,4)],\n",
    "                ['Frequency sweep G\\'\\'', round(G_dpf,4)],\n",
    "                ['Percentage difference', f\"{round(abs(G_dpt-G_dpf)/((G_dpt+G_dpf)/2),6)*100} %\"]]\n",
    "        # Set width and height of cells\n",
    "        cell_width = 60\n",
    "        cell_height = 6\n",
    "        # Loop through data and add cells to the table\n",
    "        for row in data:\n",
    "            for item in row:\n",
    "                pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "            pdf.ln()\n",
    "        pdf.ln()\n",
    "        data = [['Frequency sweep G\\' at 0.68 rad/s', round(G_pf1,4)],\n",
    "                ['Frequency sweep G\\'\\' at 0.68 rad/s', round(G_dpf1,4)],\n",
    "                ['Frequency sweep G\\' at 14.58 rad/s', f\"{round(G_pf2,4)} %\"],\n",
    "                ['Frequency sweep G\\'\\' at 14.58 rad/s', round(G_dpf2,4)]]\n",
    "        # Set width and height of cells\n",
    "        cell_width = 80\n",
    "        cell_height = 6\n",
    "        # Loop through data and add cells to the table\n",
    "        for row in data:\n",
    "            for item in row:\n",
    "                pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "            pdf.ln()\n",
    "        pdf.add_page()\n",
    "        \n",
    "        pdf.image(im5, x=pdf.get_x(), y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.image(im6, x=pdf.get_x() + 100, y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.ln(100)\n",
    "\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. HSF_{sid} before (left) and after (right)\" + \\\n",
    "                 f\" Frequency Sweep test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "\n",
    "        pdf.set_font(font, 'U', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Author's notes:\", border=0, align='J')\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=freq_text, border=0, align='J')\n",
    "        pdf.add_page()\n",
    "\n",
    "\n",
    "    if not math.isnan(eta):\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Flow step test at {test_temp} oC: \" + \\\n",
    "                       f\"The variation of viscosity (Pa s) with shear rate (1/s) at {test_temp} oC.\", border=0, align='J')\n",
    "        pdf.image(im11, x=pdf.get_x()+19, y=pdf.get_y()+5, w=0.8*pdf.w, h=0.5*pdf.w)\n",
    "        pdf.ln(110)\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. Flow step test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "\n",
    "        pdf.ln(5)\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Cross model parameters: \", border=0, align='J')\n",
    "        tnum = tnum+1\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Table {tnum}: Flow step analysis results.\")\n",
    "        pdf.set_font(font, '', 12)\n",
    "        data = [['Zero-rate viscosity (Pa s)', round(eta_0,4)],\n",
    "                ['Infinite-rate viscosity (Pa s)', round(eta_inf,4)],\n",
    "                ['Consistency (s)', round(K,4)],\n",
    "                ['Rate index', round(n,4)],\n",
    "                ['% error', round(error,4)]]\n",
    "        # Set width and height of cells\n",
    "        cell_width = 60\n",
    "        cell_height = 6\n",
    "        # Loop through data and add cells to the table\n",
    "        for row in data:\n",
    "            for item in row:\n",
    "                pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "            pdf.ln() \n",
    "        pdf.add_page()\n",
    "\n",
    "        pdf.image(im7, x=pdf.get_x(), y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.image(im8, x=pdf.get_x() + 100, y=pdf.get_y()+5, w=0.43*pdf.w, h=0.43*pdf.w)\n",
    "        pdf.ln(100)\n",
    "\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. HSF_{sid} before (left) and after (right)\" + \\\n",
    "                 f\" Flow Step test at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "\n",
    "        pdf.set_font(font, 'U', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Author's notes:\", border=0, align='J')\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=flow_text, border=0, align='J')\n",
    "        pdf.add_page()\n",
    "    \n",
    "    \n",
    "    if not math.isnan(G_pf) and not math.isnan(eta):\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Cox-Merz rule at {test_temp} oC: \" + \\\n",
    "                       f\"The viscosity variation as a function of the shear rate (1/s) and the angular frequency (rad/s) \" + \\\n",
    "                       f\"from the Flow step and Frequency sweep, respectively at {test_temp} oC.\", border=0, align='J')\n",
    "        pdf.image(im12, x=pdf.get_x()+19, y=pdf.get_y()+5, w=0.8*pdf.w, h=0.5*pdf.w)\n",
    "        pdf.ln(110)\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.cell(w=0, h=10, txt=f\"Figure {fnum}. Cox-Merz rule at {test_temp} oC.\", ln=2, align='C')\n",
    "        fnum = fnum+1\n",
    "        pdf.ln(5)\n",
    "        tnum = tnum+1\n",
    "        pdf.set_font(font, 'I', 10)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Table {tnum}: Cox-Merz rule results.\")\n",
    "        pdf.set_font(font, '', 12)\n",
    "        data = [['Flow step viscosity', round(eta,4)],\n",
    "                ['Frequency sweep |n*|', round(eta_star,4)],\n",
    "                ['Percentage difference', f\"{round(abs(eta-eta_star)/((eta+eta_star)/2), 6)*100} %\"]]\n",
    "        # Set width and height of cells\n",
    "        cell_width = 60\n",
    "        cell_height = 6\n",
    "        # Loop through data and add cells to the table\n",
    "        for row in data:\n",
    "            for item in row:\n",
    "                pdf.cell(cell_width, cell_height, str(item), border=1)\n",
    "            pdf.ln()\n",
    "        pdf.ln(5)\n",
    "        pdf.set_font(font, 'U', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=f\"Author's notes:\", border=0, align='J')\n",
    "        pdf.set_font(font, '', 12)\n",
    "        pdf.multi_cell(w=0, h=10, txt=cox_text, border=0, align='J')\n",
    "        pdf.add_page()\n",
    "\n",
    "    # Results\n",
    "    pdf.set_font(font, 'B', 12)\n",
    "    pdf.cell(w=0, h=10, txt=f\"4. Results\", ln=1)\n",
    "    pdf.set_font(font, '', 12)\n",
    "    pdf.multi_cell(w=0, h=10, txt=results_text, border=0, align='J')\n",
    "    output_filepath = \"Results/\" + filename\n",
    "    pdf.output(output_filepath, 'F')\n",
    "\n",
    "    \n",
    "    \n",
    "# Calculate deviations for certain columns\n",
    "def calculate_deviation(col1, col2, dataframe):\n",
    "    \"\"\"\n",
    "    Calculate the deviation between two columns of a given DataFrame.\n",
    "\n",
    "    Parameters:\n",
    "        col1 (str): Name of the first column.\n",
    "        col2 (str): Name of the second column.\n",
    "        dataframe (DataFrame): The DataFrame containing the columns.\n",
    "\n",
    "    Returns:\n",
    "        pandas Series: A Series containing the deviation values between the two columns.\n",
    "    \"\"\"\n",
    "    return abs(dataframe[col1] - dataframe[col2]) / ((dataframe[col1] + dataframe[col2]) / 2)\n",
    "\n",
    "\n",
    "def preprocess_data(primary_data_file=\"Data/Statistical_Analysis_Database.xlsx\"):\n",
    "    \"\"\"\n",
    "    Preprocesses the data via filtering and modifications to the primary data.\n",
    "\n",
    "    Parameters:\n",
    "        primary_data_file (str, optional): The filename of the Excel file containing the primary data. The default value is \"Data/Statistical_Analysis_Database.xlsx\".\n",
    "\n",
    "    Returns:\n",
    "        database (DataFrame): A new DataFrame with preprocessed data.\n",
    "    \"\"\"\n",
    "    # Load the excel file into a dataframe\n",
    "    database = pd.read_excel(primary_data_file, index_col=\"ID\")\n",
    "\n",
    "    # Replace NaN values of Time Sweep with values of Frequency Sweep at 3.142 rad/s\n",
    "    database[\"Time Sweep G' average (Pa)\"].fillna(database[\"Frequency Sweep G' at 3.142 rad/s (Pa)\"], inplace=True)\n",
    "    database[\"Time Sweep G'' average (Pa)\"].fillna(database[\"Frequency Sweep G'' at 3.142 rad/s (Pa)\"], inplace=True)\n",
    "    \n",
    "    # Filter values larger than threshold\n",
    "    database.loc[database[\"Cross Over Point (rad/s)\"] > 15, \"Cross Over Point (rad/s)\"] = np.nan\n",
    "    columns_to_nan = [\"Zero-rate Viscosity (Pa.s)\", \"Infinite-rate Viscosity (Pa.s)\", \"Consistency (s)\", \"Rate Index (-)\", \"Standard Error (-)\"]\n",
    "    database.loc[database[\"Zero-rate Viscosity (Pa.s)\"] > 50, columns_to_nan] = np.nan\n",
    "    database.loc[database[\"Infinite-rate Viscosity (Pa.s)\"] < 0.001, columns_to_nan] = np.nan\n",
    "\n",
    "    database[\"time/frequency G' deviation\"] = calculate_deviation(\"Time Sweep G' average (Pa)\", \"Frequency Sweep G' at 3.142 rad/s (Pa)\", database)\n",
    "    database[\"time/frequency G'' deviation\"] = calculate_deviation(\"Time Sweep G'' average (Pa)\", \"Frequency Sweep G'' at 3.142 rad/s (Pa)\", database)\n",
    "    database[\"frequency/flow step deviation\"] = calculate_deviation(\"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\", \"Flow Step Viscosity at 3.162 1/s (Pa.s)\", database)\n",
    "\n",
    "    # Filter experiments that deviate more than 20%\n",
    "    cond1 = (database[\"time/frequency G' deviation\"] > 0.2) | (database[\"time/frequency G'' deviation\"] > 0.2)\n",
    "    cond2 = (database[\"frequency/flow step deviation\"] > 0.2)\n",
    "\n",
    "    columns_to_nan_cond1 = [\"Frequency Sweep G' at 0.68 rad/s (Pa)\", \"Frequency Sweep G'' at 0.68 rad/s (Pa)\",\n",
    "                            \"Frequency Sweep G' at 3.142 rad/s (Pa)\", \"Frequency Sweep G'' at 3.142 rad/s (Pa)\",\n",
    "                            \"Frequency Sweep G' at 14.58 rad/s (Pa)\", \"Frequency Sweep G'' at 14.58 rad/s (Pa)\",\n",
    "                            \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\", \"Cross Over Point (rad/s)\"]\n",
    "    columns_to_nan_cond2 = [\"Zero-rate Viscosity (Pa.s)\", \"Infinite-rate Viscosity (Pa.s)\", \"Consistency (s)\",\n",
    "                            \"Rate Index (-)\", \"Standard Error (-)\", \"Flow Step Viscosity at 3.162 1/s (Pa.s)\"]\n",
    "\n",
    "    database.loc[cond1, columns_to_nan_cond1] = np.nan\n",
    "    database.loc[cond2, columns_to_nan_cond2] = np.nan\n",
    "\n",
    "    # Calculate tan(delta) values for certain frequencies\n",
    "    database[\"tan(delta) at 0.68 rad/s\"] = database[\"Frequency Sweep G'' at 0.68 rad/s (Pa)\"] / database[\"Frequency Sweep G' at 0.68 rad/s (Pa)\"]\n",
    "    database[\"tan(delta) at 3.142 rad/s\"] = database[\"Time Sweep G'' average (Pa)\"] / database[\"Time Sweep G' average (Pa)\"]\n",
    "    database[\"tan(delta) at 14.58 rad/s\"] = database[\"Frequency Sweep G'' at 14.58 rad/s (Pa)\"] / database[\"Frequency Sweep G' at 14.58 rad/s (Pa)\"]\n",
    "    return database\n",
    "\n",
    "\n",
    "def apply_logarithm(df, columns_to_log):\n",
    "    \"\"\"\n",
    "    Applies the logarithm (base 10) to specified columns of the given DataFrame.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame.\n",
    "\n",
    "    Returns:\n",
    "        df (DataFrame): A new DataFrame with specified columns logarithmically transformed.\n",
    "    \"\"\"\n",
    "    df[columns_to_log] = np.log10(df[columns_to_log]+1)\n",
    "    return df\n",
    "    \n",
    "\n",
    "def prepare_dataframe(df, id_vars, value_vars, replacements=None):\n",
    "    \"\"\"\n",
    "    Prepare a DataFrame for plotting by melting, dropping NaN values, and applying replacements to variable names.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame.\n",
    "        id_vars (str or list): Columns to use as identifier variables.\n",
    "        value_vars (str or list): Columns to melt and use as value variables.\n",
    "        replacements (dict, optional): Dictionary to replace variable names. Default value is None.\n",
    "\n",
    "    Returns:\n",
    "        plot (DataFrame): A new DataFrame prepared for plotting.\n",
    "    \"\"\"\n",
    "    plot = df.melt(id_vars=id_vars, value_vars=value_vars, value_name='Value', var_name='Variable')\n",
    "    plot = plot.dropna(subset=['Value'])\n",
    "    plot = plot.copy()\n",
    "\n",
    "    if replacements is not None:\n",
    "        plot[\"Variable\"] = plot[\"Variable\"].replace(replacements)\n",
    "\n",
    "    return plot\n",
    "\n",
    "\n",
    "def legend_prepare_dataframe(df, A, B):\n",
    "    \"\"\"\n",
    "    Prepare a DataFrame for legend replacement.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The input DataFrame.\n",
    "        A (str): Replacement value for 'yes'.\n",
    "        B (str): Replacement value for 'no'.\n",
    "\n",
    "    Returns:\n",
    "        DataFrame: A new DataFrame prepared for legend replacement.\n",
    "    \"\"\"\n",
    "    return df.replace({'yes': A, 'no': B})\n",
    "\n",
    "\n",
    "\n",
    "def statistics(data1, data2, column_names, alpha=0.05):\n",
    "    \"\"\"\n",
    "    Perform an independent basic statistics calculations and two-sample t-test between two sets of data for each column.\n",
    "\n",
    "    Parameters:\n",
    "        data1 (DataFrame): The first DataFrame containing the first set of data.\n",
    "        data2 (DataFrame): The second DataFrame containing the second set of data.\n",
    "        column_names (list): List of column names to perform the t-test on.\n",
    "        alpha (float, optional): Significance level for the hypothesis test. Default value is 0.05.\n",
    "\n",
    "    Returns:\n",
    "        results_df (DataFrame): A DataFrame containing the results of basic statistics and the t-test for each column.\n",
    "    \"\"\"\n",
    "    results = []\n",
    "\n",
    "    for column in column_names:\n",
    "        sample1 = data1[column].dropna()\n",
    "        sample2 = data2[column].dropna()\n",
    "\n",
    "        if len(sample1) > 0 and len(sample2) > 0:\n",
    "            count1 = len(sample1)\n",
    "            count2 = len(sample2)\n",
    "            mean1 = np.mean(sample1)\n",
    "            mean2 = np.mean(sample2)\n",
    "            std1 = np.std(sample1, ddof=1)  # Use ddof=1 for sample standard deviation\n",
    "            std2 = np.std(sample2, ddof=1)\n",
    "            max1 = np.max(sample1)\n",
    "            max2 = np.max(sample2)\n",
    "            min1 = np.min(sample1)\n",
    "            min2 = np.min(sample2)\n",
    "\n",
    "            if np.var(sample1) > 0 and np.var(sample2) > 0:\n",
    "                t_stat, p_value = stats.ttest_ind(sample1, sample2)\n",
    "                if p_value < alpha:\n",
    "                    hypothesis = 'Reject'\n",
    "                else:\n",
    "                    hypothesis = 'Accept'\n",
    "            else:\n",
    "                t_stat, p_value = np.nan, np.nan\n",
    "                hypothesis = ''\n",
    "\n",
    "            result = {\n",
    "                'Column': column,\n",
    "                'Count 1': count1,\n",
    "                'Mean 1': mean1,\n",
    "                'Std 1': std1,\n",
    "                'Max 1': max1,\n",
    "                'Min 1': min1,\n",
    "                'Count 2': count2,\n",
    "                'Mean 2': mean2,\n",
    "                'Std 2': std2,\n",
    "                'Max 2': max2,\n",
    "                'Min 2': min2,\n",
    "                't-statistic': t_stat,\n",
    "                'p-value': p_value,\n",
    "                'Hypothesis': hypothesis\n",
    "            }\n",
    "            results.append(result)\n",
    "\n",
    "    results_df = pd.DataFrame(results)\n",
    "    return results_df\n",
    "\n",
    "\n",
    "\n",
    "def categorize(df, col, limit, big_var=\"control\"):   \n",
    "    \"\"\"\n",
    "    Categorizes the data in the DataFrame based on the specified column and limit.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The DataFrame containing the data to be categorized.\n",
    "        col (str): The name of the column in the DataFrame used for categorization.\n",
    "        limit (float): The threshold value for the categorization.\n",
    "        big_var (str, optional): The variable to determine whether to consider values above or below the limit. Default value is \"control\", meaning values below the limit are considered \"positive\".\n",
    "\n",
    "    Returns:\n",
    "        df (DataFrame): The modified DataFrame with a new column \"Category\" containing the categories.\n",
    "    \"\"\"\n",
    "    df.dropna(inplace=True, subset=[col])\n",
    "    df.reset_index(drop=True, inplace=True)\n",
    "    \n",
    "    for i, row in df.iterrows():\n",
    "        if big_var == \"control\":\n",
    "            if row[\"Condition\"] == \"oa\":\n",
    "                if row[col] < limit:\n",
    "                    df.loc[i, \"Category\"] = \"TP\"\n",
    "                else:\n",
    "                    df.loc[i, \"Category\"] = \"FN\"\n",
    "            else:\n",
    "                if row[col] < limit:\n",
    "                    df.loc[i, \"Category\"] = \"FP\"\n",
    "                else:\n",
    "                    df.loc[i, \"Category\"] = \"TN\"\n",
    "        else:\n",
    "            if row[\"Condition\"] == \"oa\":\n",
    "                if row[col] > limit:\n",
    "                    df.loc[i, \"Category\"] = \"TP\"\n",
    "                else:\n",
    "                    df.loc[i, \"Category\"] = \"FN\"\n",
    "            else:\n",
    "                if row[col] > limit:\n",
    "                    df.loc[i, \"Category\"] = \"FP\"\n",
    "                else:\n",
    "                    df.loc[i, \"Category\"] = \"TN\"\n",
    "    return df\n",
    "\n",
    "\n",
    "def count_values(df):\n",
    "    \"\"\"\n",
    "    Counts the occurrences of each category in the \"Category\" column of the DataFrame.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The DataFrame containing the \"Category\" column.\n",
    "\n",
    "    Returns:\n",
    "        dict: A dictionary with the counts of each category in the format {\"Category\": count}.\n",
    "    \"\"\"\n",
    "    categories = [\"TN\", \"TP\", \"FN\", \"FP\"]\n",
    "    counts = {category: df[df[\"Category\"] == category].shape[0] for category in categories}\n",
    "    return counts\n",
    "\n",
    "\n",
    "def biomarker_validation(df, biomarkers_info):\n",
    "    \"\"\"\n",
    "    Performs biomarker validation by categorizing data and calculating various metrics.\n",
    "\n",
    "    Parameters:\n",
    "        df (DataFrame): The DataFrame containing the biomarker data.\n",
    "        biomarkers_info (list): A list of lists containing biomarker information in the format: [Biomarker_Name, Column_Name, Threshold, Category]\n",
    "\n",
    "    Returns:\n",
    "        biomarker_df (DataFrame): A DataFrame containing biomarker validation metrics for each biomarker.\n",
    "    \"\"\"\n",
    "    # Create empty Dataframe for biomarker validation \n",
    "    biomarker_df = pd.DataFrame(columns = [\"Biomarker\", \"TN\", \"TP\", \"FN\", \"FP\", \"Threshold\"])\n",
    "\n",
    "    for i in range(len(biomarkers_info)):\n",
    "        biomarkers_data = df.copy()\n",
    "\n",
    "        # Accessing elements from the nested list using integer indices\n",
    "        biomarker_name = biomarkers_info[i][1]\n",
    "        biomarker_value = biomarkers_info[i][2]\n",
    "        biomarker_category = biomarkers_info[i][3]\n",
    "\n",
    "        # Call function that categorizes the data based on the biomarker information\n",
    "        biomarkers_data = categorize(biomarkers_data, biomarker_name, biomarker_value, biomarker_category)\n",
    "\n",
    "        # Call function that counts TP, TN, FP, and FN\n",
    "        counts = count_values(biomarkers_data)\n",
    "        tn, tp, fn, fp = counts[\"TN\"], counts[\"TP\"], counts[\"FN\"], counts[\"FP\"]\n",
    "\n",
    "        # Append a row to the dataframe with specific values\n",
    "        new_row = [biomarkers_info[i][0], tn, tp, fn, fp, biomarkers_info[i][2]]\n",
    "        biomarker_df = pd.concat([biomarker_df, pd.DataFrame([new_row], columns=biomarker_df.columns)], ignore_index=True)\\\n",
    "\n",
    "    # Calculate the Sensitivity\n",
    "    biomarker_df[\"Sensitivity\"] = biomarker_df[\"TP\"] / (biomarker_df[\"TP\"] + biomarker_df[\"FN\"])\n",
    "\n",
    "    # Calculate the Specificity\n",
    "    biomarker_df[\"Specificity\"] = biomarker_df[\"TN\"] / (biomarker_df[\"TN\"] + biomarker_df[\"FP\"])\n",
    "\n",
    "    # Calculate the Accuracy\n",
    "    biomarker_df[\"Accuracy\"] = (biomarker_df[\"TP\"] + biomarker_df[\"TN\"]) / (biomarker_df[\"TP\"] + biomarker_df[\"TN\"] + biomarker_df[\"FP\"] + biomarker_df[\"FN\"])\n",
    "\n",
    "    # Calculate the Youden's Index\n",
    "    biomarker_df[\"Youden's Index\"] = biomarker_df[\"Sensitivity\"] + biomarker_df[\"Specificity\"] - 1\n",
    "   \n",
    "    return biomarker_df\n",
    "\n",
    "\n",
    "\n",
    "def create_custom_boxplot(data, out_name, x_label, y_label, x_ticklabels, hue, hue_order, legend_labels, width=0.8, dodge=True, yA=0, yB=None):\n",
    "    \"\"\"\n",
    "    Create a custom box plot using Seaborn.\n",
    "\n",
    "    Parameters:\n",
    "        data (DataFrame): The DataFrame containing the data to be plotted.\n",
    "        x_label (str): Label for the x-axis.\n",
    "        y_label (str): Label for the y-axis.\n",
    "        x_ticklabels (list): List of tick labels for the x-axis.\n",
    "        hue (str): Grouping variable for categorical coloring.\n",
    "        hue_order (list): Order of hue levels (grouping variable) for categorical coloring.\n",
    "        legend_labels (list): Labels for the legend.\n",
    "        width (float, optional): Width of the boxes. Defaults to 0.8.\n",
    "        dodge (bool, optional): When True, use a small positional jitter to avoid overlap when plotting hue levels. Default value is False.\n",
    "        yA (int, optional): The lowest y axis limit. Default value is 0.\n",
    "        yB (int, optional): The highest y axis limit. Default value is None.\n",
    "    \n",
    "    Returns:\n",
    "        None\n",
    "    \n",
    "    Notes:\n",
    "        When yB equals None, there is no change in y axis limits.\n",
    "    \"\"\"\n",
    "    \n",
    "    # Set the Seaborn theme and custom palette\n",
    "    sns.set_theme(style=\"ticks\")\n",
    "    custom_palette = ['#003366', '#E31B23', '#005CAB', '#FFC325', '#DCEEF3']\n",
    "    sns.set_palette(custom_palette)\n",
    "\n",
    "    # Create the box plot\n",
    "    fig, ax = plt.subplots(figsize=(8, 6))\n",
    "    sns.boxplot(data=data, x=\"Variable\", y=\"Value\", hue=hue, hue_order=hue_order, width=width, dodge=dodge)\n",
    "\n",
    "    # Customize the plot\n",
    "    ax.set_ylabel(y_label)\n",
    "    ax.set_xlabel(x_label)\n",
    "    ax.set_xticklabels(x_ticklabels, rotation=0)\n",
    "    if yB is not None:\n",
    "        ax.set_ylim(yA, yB)\n",
    "    \n",
    "    # Set the facecolor of the boxes to be transparent\n",
    "    for patch in ax.patches:\n",
    "        r, g, b, a = patch.get_facecolor()\n",
    "        patch.set_facecolor((r, g, b, .7))\n",
    "\n",
    "    # Add a legend with custom labels\n",
    "    handles, _ = ax.get_legend_handles_labels()\n",
    "    ax.legend(handles, legend_labels)\n",
    "    out_path = \"Results/\" + out_name\n",
    "    plt.savefig(out_path, dpi=300, bbox_inches='tight')\n",
    "    plt.show()\n",
    "\n",
    "    \n",
    "    \n",
    "def multivariate_normal_sampling(dataframe, columns, scaled_factor=1):\n",
    "    \"\"\"\n",
    "    Sample from a multivariate normal distribution based on specified columns.\n",
    "    \n",
    "    Parameters:\n",
    "    dataframe (pd.DataFrame): The input DataFrame containing the data.\n",
    "    columns (list): List of column names to sample from.\n",
    "    scaled_factor (int, optional): A scaling factor to scale the standard deviation of the original data to create a more robust model. Default value is 1.\n",
    "    \n",
    "    Returns:\n",
    "    pd.DataFrame: A DataFrame containing the sampled data.\n",
    "    \"\"\"\n",
    "    dataframe = dataframe[columns].dropna()\n",
    "    \n",
    "    # Calculate the mean vector for the specified columns\n",
    "    mean_vector = dataframe.mean().values\n",
    "    # Calculate the covariance matrix for the specified columns\n",
    "    cov_matrix = dataframe.cov().values\n",
    "    # Double the standard deviation (scale) of the covariance matrix\n",
    "    cov_matrix_scaled = cov_matrix * scaled_factor\n",
    "    \n",
    "    # Generate samples from a multivariate normal distribution\n",
    "    # using the calculated mean vector and covariance matrix\n",
    "    samples = np.random.multivariate_normal(mean_vector, cov_matrix_scaled, size=1000000)\n",
    "    # Create a new DataFrame with the sampled values and column names\n",
    "    sampled_data = pd.DataFrame(samples, columns=columns)\n",
    "    \n",
    "    return sampled_data\n",
    "\n",
    "\n",
    "\n",
    "def logistic_regression(dataframe, columns):\n",
    "    \"\"\"\n",
    "    Train, evaluate, and perform ROC analysis for a logistic regression model.\n",
    "    \n",
    "    Parameters:\n",
    "    dataframe (pd.DataFrame): The input DataFrame containing the data.\n",
    "    columns (list): List of column names to use as predictor variables.\n",
    "    \n",
    "    Returns:\n",
    "    float: The mean cross-validation accuracy of the trained model.\n",
    "    \"\"\"\n",
    "    # Define the predictor variables and the response variable\n",
    "    X = dataframe[columns]\n",
    "    y = dataframe['Condition']\n",
    "    \n",
    "    # Split the dataset into training (60%) and testing (40%) sets\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)\n",
    "    \n",
    "    # Instantiate the logistic regression model\n",
    "    log_regression = LogisticRegression()\n",
    "    \n",
    "    # Fit the model using the training data\n",
    "    log_regression.fit(X_train, y_train)\n",
    "    \n",
    "    # Perform cross-validation and calculate the mean accuracy\n",
    "    cross_val_accuracy = cross_val_score(log_regression, X_train, y_train, cv=5, scoring='accuracy').mean()\n",
    "    \n",
    "    # Calculate ROC curve metrics\n",
    "    y_pred_proba = log_regression.predict_proba(X_test)[:,1]\n",
    "    fpr, tpr, _ = metrics.roc_curve(y_test, y_pred_proba)\n",
    "    auc = metrics.roc_auc_score(y_test, y_pred_proba)\n",
    "    \n",
    "    # Create and display the ROC curve\n",
    "    plt.plot(fpr, tpr, label=\"AUC = \" + str(np.round(auc, decimals=4)))\n",
    "    plt.ylabel('Sensitivity')\n",
    "    plt.xlabel('1 - Specificity')\n",
    "    plt.legend(loc=4)\n",
    "    plt.savefig('Results/roc_auc.png', dpi=300, bbox_inches='tight')\n",
    "    plt.show()\n",
    "    \n",
    "    return cross_val_accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59757a1d",
   "metadata": {},
   "source": [
    "# Examples - How to use the different functions?\n",
    "\n",
    "## Sample Analysis Part"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "306beb8e-676a-4343-8e2a-bbb0e6529612",
   "metadata": {},
   "source": [
    "If you **do not** want to change the databases **do not** run this cell."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "85425e5b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nappend_sample_to_database(file_name=\"Data/Database.xlsx\")\\n\\nappend_samples_to_final_database(start_row_n=60, database_file=\"Data/Database.xlsx\", experiment_data_file=\"Data/Experimental_Data.xlsx\")\\n'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "append_sample_to_database(file_name=\"Data/Database.xlsx\")\n",
    "\n",
    "append_samples_to_final_database(start_row_n=60, database_file=\"Data/Database.xlsx\", experiment_data_file=\"Data/Experimental_Data.xlsx\")\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a323870-e2e8-4fbe-b5db-ba2746903109",
   "metadata": {},
   "source": [
    "Run analysis and create example report of sample 57."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a1eb8e85-ef4e-4329-a757-b46b98ff4a1c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c4533e5f502b4fa489c165bfd040e15a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(FloatRangeSlider(value=(0.0001, 100.0), description='y-axis limits:', min=0.0001, readou…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "Give a time start point (0.1, 0.2 ... 10.0):  1\n",
      "Give a time finish point (0.1, 0.2 ... 10.0):  5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------------------+----------------------+\n",
      "|      Variable      |        Value         |\n",
      "+--------------------+----------------------+\n",
      "|        Mean        | G' (Pa)     1.009489 |\n",
      "|                    | G'' (Pa)    0.752096 |\n",
      "| Standard deviation | G' (Pa)     0.002770 |\n",
      "|                    | G'' (Pa)    0.002873 |\n",
      "+--------------------+----------------------+\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "05f5fb0ff8be4ba09f8e1164c5be1531",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(FloatRangeSlider(value=(0.0001, 100.0), description='Primary y-axis limits:', min=0.0001…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Crossover point at: 1.458\n",
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "83ecd997044f41c6a27e8399c89ad515",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(FloatRangeSlider(value=(0.0001, 100.0), description='y-axis limits:', min=0.0001, readou…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8f3855cb03824690b51490a298762e93",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(FloatRangeSlider(value=(0.0001, 100.0), description='y-axis limits:', min=0.0001, readou…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------------+------------------+--------+\n",
      "|    Experiment   |     Variable     | Value  |\n",
      "+-----------------+------------------+--------+\n",
      "|    Flow step    | viscosity (Pa.s) | 0.4637 |\n",
      "| Frequency sweep |   |n*| (Pa.s)    | 0.4066 |\n",
      "+-----------------+------------------+--------+\n"
     ]
    }
   ],
   "source": [
    "analysis_results = sample_analysis(sid=\"S57\", df_ed=\"Data/Experimental_Data.xlsx\")\n",
    "\n",
    "create_sample_report(\n",
    "    snum=57, \n",
    "    database_file=\"Data/Database.xlsx\", \n",
    "    analysis_data_file=\"Data/Statistical_Analysis_Database.xlsx\", \n",
    "    txt_file=\"Data/report_57/report_template_57.txt\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beb8a618",
   "metadata": {},
   "source": [
    "## Data Analysis Part"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5330c949",
   "metadata": {
    "tags": []
   },
   "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>Column</th>\n",
       "      <th>Count 1</th>\n",
       "      <th>Mean 1</th>\n",
       "      <th>Std 1</th>\n",
       "      <th>Max 1</th>\n",
       "      <th>Min 1</th>\n",
       "      <th>Count 2</th>\n",
       "      <th>Mean 2</th>\n",
       "      <th>Std 2</th>\n",
       "      <th>Max 2</th>\n",
       "      <th>Min 2</th>\n",
       "      <th>t-statistic</th>\n",
       "      <th>p-value</th>\n",
       "      <th>Hypothesis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Frequency Sweep G' at 0.68 rad/s (Pa)</td>\n",
       "      <td>3</td>\n",
       "      <td>0.754179</td>\n",
       "      <td>0.134947</td>\n",
       "      <td>0.890700</td>\n",
       "      <td>0.620864</td>\n",
       "      <td>12</td>\n",
       "      <td>0.084015</td>\n",
       "      <td>0.090906</td>\n",
       "      <td>0.335859</td>\n",
       "      <td>0.011976</td>\n",
       "      <td>10.490686</td>\n",
       "      <td>1.029210e-07</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Frequency Sweep G'' at 0.68 rad/s (Pa)</td>\n",
       "      <td>3</td>\n",
       "      <td>0.636822</td>\n",
       "      <td>0.084670</td>\n",
       "      <td>0.711217</td>\n",
       "      <td>0.544688</td>\n",
       "      <td>12</td>\n",
       "      <td>0.100634</td>\n",
       "      <td>0.097884</td>\n",
       "      <td>0.352375</td>\n",
       "      <td>0.019062</td>\n",
       "      <td>8.655453</td>\n",
       "      <td>9.346069e-07</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Time Sweep G' average (Pa)</td>\n",
       "      <td>5</td>\n",
       "      <td>0.925918</td>\n",
       "      <td>0.090605</td>\n",
       "      <td>1.055064</td>\n",
       "      <td>0.836596</td>\n",
       "      <td>17</td>\n",
       "      <td>0.174670</td>\n",
       "      <td>0.145802</td>\n",
       "      <td>0.580591</td>\n",
       "      <td>0.008951</td>\n",
       "      <td>10.813335</td>\n",
       "      <td>8.362515e-10</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Time Sweep G'' average (Pa)</td>\n",
       "      <td>5</td>\n",
       "      <td>0.684831</td>\n",
       "      <td>0.072826</td>\n",
       "      <td>0.778752</td>\n",
       "      <td>0.617881</td>\n",
       "      <td>17</td>\n",
       "      <td>0.168383</td>\n",
       "      <td>0.130874</td>\n",
       "      <td>0.477781</td>\n",
       "      <td>0.021891</td>\n",
       "      <td>8.354783</td>\n",
       "      <td>5.924194e-08</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Frequency Sweep G' at 14.58 rad/s (Pa)</td>\n",
       "      <td>3</td>\n",
       "      <td>1.096514</td>\n",
       "      <td>0.068252</td>\n",
       "      <td>1.142702</td>\n",
       "      <td>1.018118</td>\n",
       "      <td>12</td>\n",
       "      <td>0.334869</td>\n",
       "      <td>0.202713</td>\n",
       "      <td>0.787531</td>\n",
       "      <td>0.144636</td>\n",
       "      <td>6.263553</td>\n",
       "      <td>2.908291e-05</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Frequency Sweep G'' at 14.58 rad/s (Pa)</td>\n",
       "      <td>3</td>\n",
       "      <td>0.756114</td>\n",
       "      <td>0.090646</td>\n",
       "      <td>0.822887</td>\n",
       "      <td>0.652923</td>\n",
       "      <td>12</td>\n",
       "      <td>0.236608</td>\n",
       "      <td>0.146125</td>\n",
       "      <td>0.557026</td>\n",
       "      <td>0.089976</td>\n",
       "      <td>5.788434</td>\n",
       "      <td>6.294768e-05</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)</td>\n",
       "      <td>3</td>\n",
       "      <td>0.587453</td>\n",
       "      <td>0.073280</td>\n",
       "      <td>0.647579</td>\n",
       "      <td>0.505828</td>\n",
       "      <td>12</td>\n",
       "      <td>0.091401</td>\n",
       "      <td>0.086354</td>\n",
       "      <td>0.319106</td>\n",
       "      <td>0.020403</td>\n",
       "      <td>9.097180</td>\n",
       "      <td>5.331277e-07</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Cross Over Point (rad/s)</td>\n",
       "      <td>1</td>\n",
       "      <td>0.224481</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.224481</td>\n",
       "      <td>0.224481</td>\n",
       "      <td>11</td>\n",
       "      <td>0.619009</td>\n",
       "      <td>0.246295</td>\n",
       "      <td>0.890309</td>\n",
       "      <td>0.224481</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>tan(delta) at 0.68 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>0.234911</td>\n",
       "      <td>0.024258</td>\n",
       "      <td>0.252484</td>\n",
       "      <td>0.207234</td>\n",
       "      <td>12</td>\n",
       "      <td>0.369583</td>\n",
       "      <td>0.061459</td>\n",
       "      <td>0.478311</td>\n",
       "      <td>0.294911</td>\n",
       "      <td>-3.639211</td>\n",
       "      <td>2.998486e-03</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>tan(delta) at 3.142 rad/s</td>\n",
       "      <td>5</td>\n",
       "      <td>0.180825</td>\n",
       "      <td>0.006807</td>\n",
       "      <td>0.186782</td>\n",
       "      <td>0.171379</td>\n",
       "      <td>17</td>\n",
       "      <td>0.308720</td>\n",
       "      <td>0.086018</td>\n",
       "      <td>0.541888</td>\n",
       "      <td>0.182636</td>\n",
       "      <td>-3.264944</td>\n",
       "      <td>3.875729e-03</td>\n",
       "      <td>Reject</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>tan(delta) at 14.58 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>0.148934</td>\n",
       "      <td>0.012833</td>\n",
       "      <td>0.162533</td>\n",
       "      <td>0.137036</td>\n",
       "      <td>12</td>\n",
       "      <td>0.224059</td>\n",
       "      <td>0.070364</td>\n",
       "      <td>0.323496</td>\n",
       "      <td>0.040017</td>\n",
       "      <td>-1.792700</td>\n",
       "      <td>9.630919e-02</td>\n",
       "      <td>Accept</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             Column  Count 1    Mean 1  \\\n",
       "0             Frequency Sweep G' at 0.68 rad/s (Pa)        3  0.754179   \n",
       "1            Frequency Sweep G'' at 0.68 rad/s (Pa)        3  0.636822   \n",
       "2                        Time Sweep G' average (Pa)        5  0.925918   \n",
       "3                       Time Sweep G'' average (Pa)        5  0.684831   \n",
       "4            Frequency Sweep G' at 14.58 rad/s (Pa)        3  1.096514   \n",
       "5           Frequency Sweep G'' at 14.58 rad/s (Pa)        3  0.756114   \n",
       "6   Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)        3  0.587453   \n",
       "7                          Cross Over Point (rad/s)        1  0.224481   \n",
       "8                          tan(delta) at 0.68 rad/s        3  0.234911   \n",
       "9                         tan(delta) at 3.142 rad/s        5  0.180825   \n",
       "10                        tan(delta) at 14.58 rad/s        3  0.148934   \n",
       "\n",
       "       Std 1     Max 1     Min 1  Count 2    Mean 2     Std 2     Max 2  \\\n",
       "0   0.134947  0.890700  0.620864       12  0.084015  0.090906  0.335859   \n",
       "1   0.084670  0.711217  0.544688       12  0.100634  0.097884  0.352375   \n",
       "2   0.090605  1.055064  0.836596       17  0.174670  0.145802  0.580591   \n",
       "3   0.072826  0.778752  0.617881       17  0.168383  0.130874  0.477781   \n",
       "4   0.068252  1.142702  1.018118       12  0.334869  0.202713  0.787531   \n",
       "5   0.090646  0.822887  0.652923       12  0.236608  0.146125  0.557026   \n",
       "6   0.073280  0.647579  0.505828       12  0.091401  0.086354  0.319106   \n",
       "7        NaN  0.224481  0.224481       11  0.619009  0.246295  0.890309   \n",
       "8   0.024258  0.252484  0.207234       12  0.369583  0.061459  0.478311   \n",
       "9   0.006807  0.186782  0.171379       17  0.308720  0.086018  0.541888   \n",
       "10  0.012833  0.162533  0.137036       12  0.224059  0.070364  0.323496   \n",
       "\n",
       "       Min 2  t-statistic       p-value Hypothesis  \n",
       "0   0.011976    10.490686  1.029210e-07     Reject  \n",
       "1   0.019062     8.655453  9.346069e-07     Reject  \n",
       "2   0.008951    10.813335  8.362515e-10     Reject  \n",
       "3   0.021891     8.354783  5.924194e-08     Reject  \n",
       "4   0.144636     6.263553  2.908291e-05     Reject  \n",
       "5   0.089976     5.788434  6.294768e-05     Reject  \n",
       "6   0.020403     9.097180  5.331277e-07     Reject  \n",
       "7   0.224481          NaN           NaN             \n",
       "8   0.294911    -3.639211  2.998486e-03     Reject  \n",
       "9   0.182636    -3.264944  3.875729e-03     Reject  \n",
       "10  0.040017    -1.792700  9.630919e-02     Accept  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>Biomarker</th>\n",
       "      <th>TN</th>\n",
       "      <th>TP</th>\n",
       "      <th>FN</th>\n",
       "      <th>FP</th>\n",
       "      <th>Threshold</th>\n",
       "      <th>Sensitivity</th>\n",
       "      <th>Specificity</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Youden's Index</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>G' at 0.68 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>G'' at 0.68 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.88</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>G' at 3.142 rad/s</td>\n",
       "      <td>5</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.34</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>G'' at 3.142 rad/s</td>\n",
       "      <td>5</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.58</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>G' at 14.58 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.28</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>G'' at 14.58 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Viscosity |n*| at 3.142 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.65</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>tan(delta) at 0.68 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.88</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>tan(delta) at 3.142 rad/s</td>\n",
       "      <td>5</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.65</td>\n",
       "      <td>0.941176</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>0.941176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>tan(delta) at 14.58 rad/s</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>0.916667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Biomarker TN  TP FN FP  Threshold Sensitivity  \\\n",
       "0               G' at 0.68 rad/s  3  12  0  0       2.18         1.0   \n",
       "1              G'' at 0.68 rad/s  3  12  0  0       1.88         1.0   \n",
       "2              G' at 3.142 rad/s  5  17  0  0       4.34         1.0   \n",
       "3             G'' at 3.142 rad/s  5  17  0  0       2.58         1.0   \n",
       "4              G' at 14.58 rad/s  3  12  0  0       7.28         1.0   \n",
       "5             G'' at 14.58 rad/s  3  12  0  0       3.06         1.0   \n",
       "6  Viscosity |n*| at 3.142 rad/s  3  12  0  0       1.65         1.0   \n",
       "7       tan(delta) at 0.68 rad/s  3  12  0  0       0.88         1.0   \n",
       "8      tan(delta) at 3.142 rad/s  5  16  1  0       0.65    0.941176   \n",
       "9      tan(delta) at 14.58 rad/s  3  11  1  0       0.48    0.916667   \n",
       "\n",
       "  Specificity  Accuracy Youden's Index  \n",
       "0         1.0       1.0            1.0  \n",
       "1         1.0       1.0            1.0  \n",
       "2         1.0       1.0            1.0  \n",
       "3         1.0       1.0            1.0  \n",
       "4         1.0       1.0            1.0  \n",
       "5         1.0       1.0            1.0  \n",
       "6         1.0       1.0            1.0  \n",
       "7         1.0       1.0            1.0  \n",
       "8         1.0  0.954545       0.941176  \n",
       "9         1.0  0.933333       0.916667  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArIAAAH9CAYAAADmn5+kAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABb40lEQVR4nO3dd3iT1f//8VdaVutgCRRBhkDZ0EJYgmxQUFGGH1RARGR+cIEMERmKgIKDLSqC4gArUEFRkaqsDzIKzgpIKbILsrGFrvP7g1/zbehKaUJyt8/HdXFd5c6dk3dz905eOTnn3DZjjBEAAABgMX7eLgAAAAC4FgRZAAAAWBJBFgAAAJZEkAUAAIAlEWQBAABgSQRZAAAAWBJBFgAAAJZEkAUAAIAlFfB2Adeb3W5XQkKCSpUq5e1SAAAAkIGTJ0+qUKFC2rFjR5b75bsge/nyZSUnJ3u7DAAAAGQiKSlJrlx8Nt8F2dKlS0uSIiIivFwJAAAAMtKuXTuX9mOMLAAAACyJIAsAAABLIsgCAADAkgiyAAAAsKR8N9kLAADkfcnJyUpMTPR2GchAwYIF5e/v75a2CLIAACDPMMbo+PHjOnfunEvLN+H6s9lsKlq0qIKCgmSz2XLVFkEWAADkGefOndPZs2dVqlQp3XDDDbkOSnAvY4z+/fdfnTx5UgEBASpWrFiu2iPIAgCAPMEYoxMnTujmm2/WLbfc4u1ykImAgABdvnxZJ06cUNGiRXP1YYPJXgAAIE9ITk5WcnKybr75Zm+XgmzcfPPNjuOVGwRZAACQJyQlJUmSChTgC2dfl3qMUo/ZtSLIAgCAPIVxsb7PXceIjywAACDPS05OVkpKilce28/Pz23LTcEZQRYAAORpycnJatK0qf45edIrj39LqVLa+tNPOQqzffr0Ubly5TRt2rR0t40ZM0ZHjhzRkiVL3FJf9erVNXXqVHXr1k3GGIWHh6tly5YqWbKkVqxYoeeff1579uxxy2O5G0EWAADkaSkpKfrn5Ek1vG+gbH7Xd1SlSUlR5Op3lJKSYole2e3bt2vMmDGKiIjwdikuIcgCAIB8webnJz+/6xsmvTOY4dpZ7SISTPYCAACwqAsXLujFF19U06ZN1bBhQz366KP67bffHLcbY/Tee++pU6dOqlOnjho2bKhBgwbp0KFD6draunWrHn30UUlSu3bttGLFCsdtK1asUIcOHVS3bl1169ZNv/zyiyRp8eLFCg0NVXx8vGPflJQUtWzZUh9++KGnfm0HgiwAAIAFGWM0YMAAHThwQAsWLNBnn32mkJAQPfzww4qKipIkffDBB1qwYIFGjhypb7/9VvPmzVNMTEyGY29DQ0M1e/ZsSVJYWJg6d+7suG3p0qV6/fXXtXz5chUqVEjPPPOMJKlLly5KTEzU2rVrHfv+73//0+nTp3Xvvfd68Le/gqEFAABYiDHGqfcrq/0k15Y5CggIYMkqH7R69Wp9++236bYnJCSoQYMG+umnn7Rr1y5t2bJFJUqUkCQNHz5cO3fu1Icffqhp06apQoUKmjZtmtq2bStJKleunDp16qSvvvoqXbuFChVS0aJFJUklSpRQkSJFHLdNmTJFVatWlST1799fw4YN06lTp1SyZEm1bdtWq1at0v333y9JWrlypdq2beuoyZMIsgAAWIQxRj169FBkZKRb27Xb7QoLCyPM+pi2bdvqueeeS7d9xowZOnv2rP744w9JV4YBpJWQkKDLly872vjll180a9Ys/f3334qOjtZff/2lMmXK5KiWypUrO35OvXLapUuXJEndu3fX4MGDFRsbqxtuuEHr1q3TzJkzc9T+tSLIAgBgIYTN/OOGG25QxYoVM9x+9uxZpaSk6MYbb3Qay5qqUKFCkqR3331Xs2fPVrdu3dS4cWP16dNHERERGfbIZiWjFRdSe/1btGihUqVK6auvvlKxYsV000036c4778xR+9eKIAsAgEXYbDaFhYVlO7QgLi5OdrtdkrRjxw4FBgZmuT9DC6wpODhYFy9eVEJCgqpVq+bYPm7cONWoUUO9e/fW/PnzNWzYMA0cONBx+8KFCzNdneBa/g78/f31wAMPaO3atSpWrJjuv//+67bUGEEWAAALsdls2QbTtAIDA3O0P6zjzjvvVM2aNfXMM89o3LhxuvXWW7V06VItX75c77//viSpbNmy2rx5s9q2bSs/Pz998cUXWrt2rW655ZYM20z9W9m9e7eKFy/uci3du3fXu+++q4IFC2rkyJG5/+VcxKoFAAAgXzApKUpJSb6u/4wHL4vr7++v999/X/Xq1dOzzz6rLl26aOvWrZo9e7aaNWsmSXrttdd06dIlde/eXb1799bevXs1adIknTp1SocPH07XZnBwsFq1aqVnnnlGy5Ytc7mWihUrKiQkRDVr1lSVKlXc9jtmx2astvJtLqUOiLbKFSsAAMipuLg41apVS5IUFRWVb3pkL126pJiYGFWuXNlpxr0VL1FrNcYYdezYUQMHDtSDDz6Y7f6ZHatUruY1hhYAAOBlri6p5aq4uLgMf3YHK46n9ff319afflKKB3tHs+Ln55dnQ2xiYqK+//57/fTTT7p48aLuueee6/r4BFkAALwsPj7e0YPqbqmTvtzFqj28/v7+eTZMelPBggU1efJkSdL06dOv+98GQRYAAADXbOPGjV57bIIsAAA+JLRzP/n5F8x1Ozm5sld2UpITtWvNoly3A7gbQRYAAB/i519Q/gVyH2SB/IDltwAAAGBJBFkAAABYEkMLAADwIclJid4uIR1frAmQCLIAAHhd2msT/fy1b0+qymfXUYKPI8gCAIA8Lzk52ZIXREhOTtayZcu0YsUKRUdHy9/fX1WrVlXPnj31wAMPpFuV4sCBA7rrrrtUs2ZNhYeHu6F630aQBQDAy9KGkZBO/Xxu1YLkpERHT7HVruolXQmDzZs21XEvXaI2qFQpbb6GS9QmJSVp6NCh+u233zRs2DA1b95cycnJ2rx5s6ZMmaKIiAjNnDnTqd0VK1aocuXK+vPPP/Xzzz8rJCTEzb+NbyHIAgDgQ/wLsPyWu6WkpOj4yZNadV83FfC7vvPck1JS1GX1CqWkpOQ4yL799tuKjIzUihUrVLFiRcf2KlWqqHHjxurRo4cWLlyogQMHSroS2MPDw9W7d2+tWrVKS5cuzfNBllULAABAvlDAz88r/66FMUYfffSRunbt6hRiU9WoUUP333+/lixZ4hgysWnTJsXGxuqOO+7Q3XffrTVr1ujcuXO5es58HUEWAADAx8TExOjMmTNq0KBBpvs0a9ZMJ06c0OHDhyVJy5cvV/ny5VWnTh117txZly9f1sqVK69XyV5BkAUAAPAxZ8+elSQVL148031Sbzt9+rTOnj2r77//Xp07d5Yk3X777apVq5aWLl3q8Vq9iSALAADgY4oVKyZJunDhQqb7pA4bKF68uFatWqXExERHkJWkzp07KyYmRlu2bPFord7EZC8AAAAfU7FiRZUqVUrbtm1Tx44dM9xn69atKlWqlMqXL68VK1ZIkrp37+64PXXN36VLl6pZs2aeL9oLCLIAAAA+xt/fX48++qjmzZunnj17qlq1ak637969W+Hh4Ro0aJD27t2rP//8U4MHD9Y999zjtN9rr72miIgInTx5UqVKlbqev8J1wdACAAAAH9S/f3/deeed6t27tz7++GP9/fff+vvvv/Xxxx+rb9++atKkiQYOHKjly5crICBAjz/+uIKDg53+DRo0SImJifr888+9/et4hE/1yM6bN09btmzRkiVLHNu+//57zZ07V/v371fx4sV111136emnn1aRIkW8WCkAALCaJC9c2Ss3j+nv769Zs2ZpxYoVCgsL05tvviljjKpVq6bnnntOPXr0UGJiolavXq377rtPRYsWTddGo0aNVK9ePYWFhWnQoEHyu87r6HqazwTZxYsXa9asWWrUqJFj244dOzRs2DA988wzuuuuu/T3339r/PjxOnv2rKZOnerFagEAgFX4+fkpqFQpdVm9wiuPH1Sq1DUHSJvNpu7duzuNfU2rUKFC2rp1a5ZthIWFXdNjW4HXg2xsbKxeeOEFRUZGqnLlyk63LV26VE2bNnVcsaJixYp69tlnNXbsWE2aNEmFChXyRskAAMBC/P39tfmnnxwXDrje/Pz8cnxVL7jG60H2jz/+UNGiRbVq1SrNnTtXR44ccdz2+OOPZ/gJJikpSRcvXlSJEiWuZ6kAAMCi/P39CZN5kNeDbNu2bdW2bdsMb6tVq5bT/xMSErRo0SLVrl07yxDbrl27TG87duyYypYte23FAgAAwGd4Pci6KikpSaNGjdK+ffv08ccfe7scAAAAeJklguzFixf1zDPPaOvWrZo1a5bq16+f5f4RERGZ3pZVby0AAACsw+eD7IkTJzRgwAAdPnxY7777rpo2bertkgAAgA9LvaIVfJe7jpFPB9lz586pb9++unjxoj755BNVr17d2yUBAAAfVbBgQUlSXFycAgICvFwNshIXFyfp/47ZtfLpIDt16lQdOnRI7733nkqUKKGTJ086bitRogSzDwEAgIO/v7+KFSumEydOSJICAwNls9m8XBXSMsYoLi5OJ06cULFixXKd5Xw2yKakpGjNmjVKTExU3759090eERGh8uXLe6EyAADgq4KCgiTJEWbhm4oVK+Y4VrnhU0F22rRpjp/9/Pz066+/erEaAABgNTabTWXLllXp0qWVmJjo7XKQgYIFC7rtW3WfCrIAAADuwAUQ8odru/AvAAAA4GUEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkAQAAYEkFvF0AAAD4PynJiW5pxxgjSbLZbLluy101Ae5GkAUAwIfsWrPI2yUAlsHQAgAAAFgSPbIAAHhZQECAoqKi3NZeXFyc7Ha7JGnHjh0KDAx0W9sBAQFuawvILYIsAABeZrPZ3Bo20woMDPRY24C3MbQAAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCUV8HYBAPInY4zi4+Nd2k+SbDabS+0GBAS4vC8AwNoIsgCuO2OMevToocjISLe3bbfbFRYWRpgFgHyAoQUAvIKgCQDILXpkAVx3NptNYWFh2Q4tiIuLk91ulyTt2LFDgYGB2bbN0AIAyD8IsgC8wmazuRRMUwUGBuZofwBA3sfQAgAAAFgSQRYAAACWRJAFAACAJflUkJ03b5769OnjtO3PP/9U7969FRISotatW2vhwoVeqg4AAAC+xGeC7OLFizVr1iynbWfOnFG/fv1UqVIlLV++XE8++aRmzpyp5cuXe6lKAAAA+Aqvr1oQGxurF154QZGRkapcubLTbZ999pkKFSqkiRMnqkCBAqpSpYr+/vtvvfvuu+revbuXKgYAAIAv8HqP7B9//KGiRYtq1apVql+/vtNtO3bsUKNGjVSgwP/l7aZNmyomJkanTp263qUCAADAh3i9R7Zt27Zq27ZthrcdP35cwcHBTttKly4tSTp69KhKliyZ4f3atWuX6eMdO3ZMZcuWvcZqAQAA4Cu83iOblUuXLqlQoUJO2woXLixJunz5sjdKAgAAgI/weo9sVooUKaKEhASnbakBNqsr/ERERGR6W1a9tQAAALAOn+6RDQoK0okTJ5y2pf6/TJky3igJAAAAPsKng2yjRo0UGRmp5ORkx7YtW7aocuXKmY6PBQAAQP7g00G2e/fuunjxol544QXt27dPK1as0AcffKBBgwZ5uzQAAAB4mU8H2ZIlS+q9995TTEyMunbtqjlz5mjUqFHq2rWrt0sDAACAl/nUZK9p06al21avXj0tW7bMC9UAAADAl/l0jywAAACQGYIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALMmnLlELAACyZoxRfHx8lvvExcVl+HNmAgICZLPZcl0bcL0RZAEAsAhjjHr06KHIyEiX72O3213aJywsjDALy2FoAQAAFkLYBP4PPbIAAFiEzWZTWFhYtkMLpCu9t6n3yQ5DC2BVBFkAACzEZrMpMDDQ22UAPoGhBQAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsiSALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsqYC3CwCQdxhjFB8f77b24uLiMvzZXQICAmSz2dzeLgDg+iDIAnCb+Ph41apVyyNt2+12t7cZFRWlwMBAt7cLALg+GFoAAAAAS6JHFoBHhHbuJz//grluxxgjSW4bApCSnKhdaxa5pS0AgHcRZAF4hJ9/QfkXyH2QBQAgMwwtAAAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYkiWCbGJiot588021bt1aoaGheuSRR7Rz505vlwUAAAAvskSQnT9/vpYvX67JkycrPDxct99+uwYMGKDY2FhvlwYAAAAvsUSQjYiI0L333qsWLVqoYsWKGjNmjC5evKiff/7Z26UBAADAS3J8idqkpCRt27ZNW7Zs0eHDh3XhwgUVL15ct956q1q2bKkGDRq47ZroqYoVK6YffvhBvXv3VtmyZbVs2TIVKlRINWvWdOvjAAAAwDpcDrIJCQn69NNPtWjRIh0/flxFixbVrbfeqoCAAB0/flw//vijFixYoNKlS2vAgAHq2bOnChUq5JYiX3jhBT377LNq166d/P395efnp5kzZ6pChQoZ7t+uXbtM2zp27JjKli3rlroAAADgPS4F2V9//VWjR4+Wv7+/Hn74YXXq1CnDELlnzx6tX79eS5Ys0YcffqjXXntNoaGhuS4yOjpaN998s+bOnasyZcooLCxMo0eP1kcffaQaNWrkun0AAABYj0tBduTIkXruuefUoUOHLPerXr26qlevroEDB2rNmjUaM2aMvv3221wVeOTIEY0cOVKLFy+W3W6XJNWtW1f79u3T7NmzNXfu3HT3iYiIyLS9rHprAQAAYB0uBdnVq1fneJhA586d1b59+2sqKq1ff/1ViYmJqlu3rtP2+vXra8OGDbluHwAAANbk0qoFOQmxxphrul9mUsez7tmzx2n73r17VbFixVy3DwAAAGvK8aoFkvTVV19p27ZtSkxMdARXY4zi4uL0888/u7WntF69erLb7Ro9erQmTJigoKAghYeHa8uWLfrkk0/c9jgAAACwlhwH2Tlz5mjOnDm66aablJSUpIIFC6pAgQI6ffq0/Pz89OCDD7q1QD8/P82bN09vvfWWnn/+eZ07d07BwcFavHixQkJC3PpYAAAAsI4cXxBh5cqV6tKli7Zt26bHHntMbdq00f/+9z99/vnnKlasmKpVq+b2IosWLaoJEybohx9+0M6dO7V06VI1btzY7Y8DAAAA68hxkI2NjdX9998vm82m2rVra9euXZKkOnXqaPDgwQoLC3N7kQAAAMDVchxkAwMDHVfuqlSpkg4fPqxLly5JkmrWrKnDhw+7t0IAAAAgAzkOsnXr1tXKlSslSRUqVJC/v7/+97//Sbpy4QJ3Xc0LAAAAyEqOJ3sNHjxY/fr104ULF/T222+rS5cuGjNmjJo0aaJNmza5Ze1YAAAAIDs5DrKNGjXS559/7ljXdfz48fLz89POnTt19913a8yYMW4vEgAAALjaNa0jW6NGDdWoUUOSVLhwYb388stuLQoAAADIjstBdu/evfrkk0909OhRVahQQQ899JCqVq3qydoAAACATLkUZHfu3Km+ffsqOTlZxYsX16ZNm7R06VLNmDFDd999t6drBAAAANJxadWCuXPnqkqVKvruu++0efNmbd68WY0aNdL06dM9XR8AAACQIZeC7G+//aZhw4apXLlykqTixYtr5MiROnr0qE6cOOHRAgEAAICMuBRkL168qFtuucVpW6VKlWSM0ZkzZzxSGAAAAJAVl4JsSkqK/Pycdy1cuLAkKSkpyf1VAQAAANnI8ZW9AAAAAF/g8vJbJ0+e1NGjRx3/T05OliT9888/Ttsl6dZbb3VTeQAAAEDGXA6yw4YNy3D74MGD0237888/r70iAAAAwAUuBdmpU6d6ug4AAAAgR1wKsl27dvV0HQAAAECOuDTZa+bMmbp8+XKOGo6Li9Obb755TUUBAAAA2XEpyJ4/f14dO3bU+++/r9jY2Cz3/eeffzR//nzdddddOn/+vFuKBAAAAK7m0tCCF198UR06dNC0adM0Y8YM1a9fX/Xq1VP58uUVEBCgCxcu6NixY9q5c6d2796tqlWrasqUKbrzzjs9XT8AAADyKZdXLWjatKnCw8P1448/avXq1fryyy916tQpx+233HKLWrRoof/+979q06aNR4oFAAAAUrkcZFO1bt1arVu3liRdunRJ58+fV7FixVSoUCF31wbAwpKTEr1dQoZ8tS4AQM7lOMimVaRIERUpUsRdtQCwOGOM4+efv17kxUpck7ZeAID1cIlaAAAAWFKuemQBIC2bzeb4OaRTP/kXKOjFajKWnJTo6C1OWy8AwHoIsgA8wr9AQZ8MsgCAvIOhBQAAALAkl3pk58yZI5vNpjJlyqh69eqqW7eup+sCAAAAsuRykJUkPz8/VahQQd98841HiwIAAACy41KQjYiIkM1m0y233MJ6sQAAAPAJLgXZ7du3S5KCgoJUvXp1FS9e3KNFAQAAANlxKciOGTNGNptNxhgVKVJEP//8s4fLshZjjOLj413aT3JtyZ+AgACWBgIAAMiCS0H2ww8/lM1mU6lSpVSxYkVP12Qpxhj16NFDkZGRbm3XbrcrLCyMMAsAAJAJl4Js48aNPV2HpRE2AQAArj+XL4iQkJCgxYsX64YbblCvXr2UnJysOnXqOO3TpUsXvfrqq24v0pfZbDaFhYVlO7QgLi5OdrtdkrRjxw4FBgZmuT9DCwAAALLmUpCNj49X7969tXv3bvXt29ex3Rij//znPypdurSio6O1atUq9evXTzVq1PBYwb7IZrNlG0zTCgwMzNH+AAAASM+lILtkyRLFxMTo008/Vb169Zxu69mzp2rXrq2UlBT9+eef+uyzzzR+/HiPFAsAAACkcukStd9884369OmTLsSm/erbz89P3bt315YtW9xbIQAAAJABl4JsTEyMmjZtmm576nJSqerUqaNjx465pzIAAAAgCy4NLbDZbPLzc868/v7++uOPP+Tv7+/YlpKSwpW/AAAAcF241CNbrlw57dmzJ932tCFWkn7//XeVL1/ePZUBAAAAWXApyLZq1Uoff/yxLl++nOk+Fy9e1LJly9ShQwe3FQcAAABkxqUg26dPH509e1aPPvqooqOj091++PBhDR48WImJiXr44YfdXiQAAABwNZfGyJYpU0azZs3S8OHDde+996p69eqqVKmSJOnIkSP6448/VLx4cc2ePVvFihXzYLkAAADAFS5f2atJkyZavXq1PvnkE33//ffatGmTUlJSVK5cOQ0cOFC9e/fWLbfc4slaAQAAAAeXg6wklShRQsOGDdOwYcM8VQ8AAADgEpfGyAIAAAC+hiALAAAASyLIAgAAwJIIsgAAALAkgiwAAAAsyaVVC7Zv356jRhs1anRNxQAAAACucinI9unTRzabLcPbjDGS5HT7n3/+6YbSAFhZSnKiW9rJ6DUmN9xVFwDA+1wKsh9++KHj56NHj+rFF19U9+7d1alTJ5UqVUpnz57V999/r6VLl+qll17ySKHh4eF65513dOjQIVWoUEHDhg1Tp06dPPJYAHJv15pF3i4BAJDHuRRkGzdu7Pi5T58+euyxxzRixAinfRo0aKAiRYpo0aJF6ty5s1uL/OKLLzR27FiNHj1arVu31pdffqnhw4crKChIoaGhbn0sAAAAWEOOruwlSb/++quGDBmS4W2hoaF69913c11UWsYYzZw5U3379lXfvn0lSf/973+1c+dObdu2jSAL+JCAgABFRUW5rb24uDjZ7XZJ0o4dOxQYGOi2tqUr9QIArCvHQTYoKEg//vij7rjjjnS3ffPNN6pQoYJbCku1f/9+HTlyRPfdd5/T9oULF7r1cQDkns1mc3vYTBUYGOixtgEA1pTjINuvXz9NnDhRJ0+eVNu2bVWiRAn9888/+uabb/Tjjz/qjTfecGuBBw4ckHSlZ6Z///6KiopS+fLlNWTIELVt29atjwUAAADryHGQfeihh5SUlKT58+fr66+/dmwvW7asZsyY4fYJWBcvXpQkjR49WsOGDdNzzz2nb7/9VkOHDtWiRYvUrFmzdPdp165dpu0dO3ZMZcuWdWuNAAAAuP5yHGQlqXfv3urdu7eio6N1/vx5FS9eXJUqVXJzaVcULFhQktS/f3917dpVklSzZk1FRUVlGmQBAACQ911TkE1VpUoVd9WRqaCgIElScHCw0/aqVavqxx9/zPA+ERERmbaXVW8tAAAArCPHQfb06dN65ZVX9OOPPyo+Pt6xWHkqm83m1lnLtWrV0g033KBffvnFMXtZkvbu3ev2iWUArh9jjOLj47PcJy4uLsOfsxIQEOC2iycAAHxbjoPsxIkTtX79et1zzz0KCgqSn5+fJ+pyKFKkiJ544gnNnTtXZcqUUb169fTVV19p8+bNWrx4sUcfG4BnGGPUo0cPRUZGunyftB9ks9svLCyMMAsA+UCOg+zGjRs1duxY9ezZ0xP1ZGjo0KEKCAjQm2++qdjYWFWpUkWzZ89WkyZNrlsNANyLoAkAyK0cB9lChQrptttu80QtWerXr5/69et33R8XgPvZbDaFhYVlO7RAkmP4kqvBl6EFAJB/5DjIdujQQV9++WWGF0QAAFd58uIJAID8IcdBtlatWnrrrbd06NAh1a9fX0WKFHG63Waz6b///a/bCvQGVyah5NS1TFpxFT1QAAAgP7KZq5cdyEaNGjWybtBm059//pmrojwpdfmtrJboiouLU61ata5XSbkWFRVFzxYAAMgzXMlr0jX0yO7evfvaKgIAAMjDXP1Gl7H/7pOrCyJk5MKFC7rpppvc3azXhHbuJz//gm5pK6d/uFlJSU7UrjWLct0OAADIvWtZVtBVLCuYuRwH2YSEBC1evFjbtm1TYmKiI5wZYxQXF6d9+/bpl19+cXuh3uLnX1D+BdwTZAEAQN5F0Lz+chxkX3vtNX300UcKDg7W6dOnVbhwYZUoUUJ79+5VYmKihg0b5ok6AQAAfJarywrGxcU5LvCyY8cOl+a4MLQgczkOsmvXrtVjjz2mMWPGaMGCBYqKitLMmTMVGxur3r17KyUlxRN1AgAA+LScLisYGBjIZO1cyvH1ZU+fPq1WrVpJkqpXr67ffvtNklSmTBkNHDhQa9ascW+FAAAAQAZyHGRvuukmJSQkSJIqVaqkY8eO6eLFi07/BwAAADwtx0ML7Ha7lixZokaNGql8+fIKCAjQd999p65du2rXrl268cYbPVEnAADAdefuiyR58gJJUv4bT5vjIDts2DD16tVLgwYN0pIlS/TII49o/PjxWrJkifbs2aOHH37YE3UCAABcd/Hx8R67SFLqpC93ym8XScpxkK1evbq+/vpr7d27V5I0YsQI3Xjjjdq5c6fatm2rgQMHur1IAAAA4GrXdEGEUqVKqVSpUpKuzNAbPHiwW4sCAADwNe66SJI7L5Ak5e+LJLn9yl4AAAB5ERdJ8j05XrUAAAAA8AUEWQAAAFgSQRYAAHjFunXr1Lx5c61bt87bpcCiXBoju337dtlsNpUpU0a33Xabp2sCAAB5XHx8vMaNG6fjx49r3Lhxat68uQICArxdVpaSkxK9XUKGfLWu68GlINunTx/ZbDYZYxQYGKidO3d6ui4AAJCHzZs3T7GxsZKk2NhYzZ8/X8OHD/dyVemlrjAgST9/7fsrA6StNz9wKchOnTpVNptNpUqVUrVq1TxdEwAAyMMOHDig+fPnO0KXMUbz5s1Tt27dVKlSJe8WB0txKch27drV03UAAIB8wBij8ePHp+s5TN3+wQcf+NQlVtPWEtKpn08uv5WclOjoLfal5+56yNE6smvWrJHNZlOnTp2UkpKiDh06ON1+33336ZlnnnFnfQAAIA+Jjo7Whg0b0m1PTk7Whg0bFB0drapVq3qhsuz5F2AdWV/j0qoFSUlJGjx4sEaMGOH44zPG6MiRI6pWrZoaN26soKAgvffeezp48KBHCwYAANZVpUoVtWzZUv7+/k7b/f391apVK1WpUsVLlcGKXOqRXbZsmTZv3qyZM2eqY8eOTrc9+eSTql27ti5duqS77rpLS5cu1ahRozxSLADg+jDGKD4+3qX9JNe+zgwICMh3X3siPZvNppdeeknt27fPcDt/I8gJl4Ls6tWr1bNnz3QhNq0iRYqoe/fuioiIcFtxAIDrzxijHj16KDIy0q3t2u12hYWFEVSgSpUqaciQIZozZ46MMbLZbBo6dKgqVqzo7dJgMS4NLdi3b59atmyZ7X4NGjRgaAEA5AGETXja0KFDVaZMGUlSUFCQhgwZ4uWKYEUu9cgmJSWlW6TY399fa9euVVBQkNM2Pz8uFgYAVmaz2RQWFpbt0IK4uDjZ7XZJ0o4dOxQYGJjl/gwtQFoBAQGaPHmyJkyYoEmTJvn8xRDgm1wKsmXKlFFMTIwaNWrktL1ChQpO/9+7d69uvfVW91UHAPAKm82WbTBNKzAwMEf7A5LUvn37dGNlgZxwqfu0RYsWWrp0qVJSUjLdJzExUZ999pnatGnjtuIAAACAzLgUZHv16qX9+/frqaee0pkzZ9LdHhcXp5EjR+rEiRN6+OGH3V4kAAAAcDWXhhbcfvvtmjJlisaOHat27dqpWbNmjkvIHTlyRJs2bVJycrKmT5+usmXLerJeAAAAQFIOruzVuXNn1ahRQ++++66+//57xzJbgYGBateunQYNGuSzV+IAAABA3pOjS9Tefvvtmjp1qiTp/PnzSklJUbFixTxRFwAAAJClHAXZtG6++WZ31gEAAADkCIu+AgAAwJKuuUc2v0hOSvR2CRny1boAAACuF4JsBowxjp9//nqRFytxTdp6AQAA8guGFgAAAMCS6JHNQNprgYd06if/AgW9WE3GkpMSHb3FXLscQHaMMYqPj3drm3FxcRn+7A4BAQG8tgHIFkE2G/4FCvpkkAWAnIiPj1etWrU81r7dbndre1FRUQoMDHRrmwDyHoYWAAAAwJLokQWAfCa0cz/5+bvnm6bUyabuGAaQkpyoXWt8f4ItAN9BkAWAfMbPnyFTAPIGhhYAAADAkgiyAAAAsCSGFgAAALggJdk9V9V059hyyX11WRFBFgAAwAVMRvQ9DC0AAACAJdEjCwAAkImAgABFRUW5rb24uDjHBUR27Njh9gt/BAQEuLU9X0eQBQAAyITNZvPYVeYCAwO5gl0uMbQAAAAAlkSPLPI0Y4zi4+Nd2k9ybQZpQECA22aaAgCAa0eQRZ5ljFGPHj0UGRnp1nbtdrvCwsIIswAAeJmlhhbExMQoNDRUK1as8HYpsAjCJgAAeZdlemQTExP13HPPKS4uztulwCJsNpvCwsKyHVqQ0xmkDC0AAMA3WCbIzp49WzfccIO3y4DF5HS2KTNIAQCwDksMLdi+fbuWLVumV1991dulAAAAwEf4fI/s+fPnNWrUKI0bN05ly5Z16T7t2rXL9LZjx4653A4AAAB8l8/3yE6cOFEhISG67777vF0KAAAAfIhP98iGh4drx44dWr16dY7uFxERkeltWfXWZiQlOTFH+2clJ2uVZseddQEAAFiRTwfZ5cuX69SpU2rdurXT9gkTJmjhwoX66quvPF7DrjWLPP4YAAAAyDmfDrIzZszQpUuXnLZ17NhRTz31lDp37uylqgAAAOALfDrIlilTJsPtJUuWVLly5Tz2uAEBAYqKinJrmzldqzQnAgIC3NaWFbh62VlXpV2b2N3rFLPmLAAAnuPTQdZbcrr2aE6xVmnuxMfHq1atWh5pO/XDhrtERUVxrAEA8BDLBdk9e/Z4uwQAAAD4AMsFWSCt0M795OdfMNftuHtFCSYJAgDgeQRZWJqff0H5F8h9kAUAANbj8xdEAAAAADJCkAUAAIAlEWQBAABgSQRZAAAAWBKTvQAgn0lOSvR2CRny1boA+C6CLCzNF9/4fLEmIHWJOUn6+WvfXx4ubb2AVbhy5clruZokV4nMHEEWlmOlN2TejAEgfzDGqEePHoqMjHT5Pq5eTdJutyssLIwwmwGCLADkA2nfAEM69fPJ9ZeTkxIdH055w4YV8Xd7/RFkYTm+/obMmzF8nX8BLiQCuJvNZlNYWFi2QwuknF9NkqEFmSPIwtJ4QwYA+AqbzabAwEBvl5GvsPwWAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJCZ7AUA+k5Lsvot25HT2dVbcWReA/IEgCwD5zK41vn0hEQBwFUHWDTxxSTrWjAMAAMgaQTaXPHVJOi5HB8CdAgICFBUV5fL+rnxAj4+P15133ilJ2rhxowICArKtIScLwANAdgiybkDYBODrcrJQ+7V8QE8NtFnhAzoAdyPI5pKnLknH0AIA3sTrDwArIMi6AZekA5CX8AEdgFUQZGFp7lquhyWEAGd8QAdgBQRZWBrLCAEAkH9xZS8AAABYEj2ysJycLiOUnbi4OMeSaDt27HDr16ksIQQAgOcQZGE5nhy7FxgYyLhAXJN169ZpwoQJmjRpktq3b+/tcgAgX2BoAQDkUnx8vMaNG6cjR45o3LhxLs32BwDkHkEWAHJp3rx5io2NlSTFxsZq/vz5Xq4IAPIHgiwA5MKBAwc0f/58xxJuxhjNmzdPBw4c8G5hAJAPEGQB4BoZYzR+/HhHiM1uOwDAvQiyAHCNoqOjtWHDBiUnJzttT05O1oYNGxQdHe2lygAgfyDIAsA1qlKlilq2bCl/f3+n7f7+/mrVqpWqVKnipcoAIPfWrVun5s2ba926dd4uJVMsvwUA18hms+mll15Kt9xW6nZ3XPIYsCJjjEurd+T08uABAQGcV9dJ6mosx48f17hx49S8eXOfXBudIAsAuVCpUiUNGTJEc+bMkTFGNptNQ4cOVcWKFb1dGuAVxhj16NFDkZGRbm/bbrcrLCyMMHsdZLQay/Dhw71cVXoMLQCAXBo6dKjKlCkjSQoKCtKQIUO8XBHgXQRNa7PSaiz0yAJALgUEBGjy5MmOK3v54tdvwPVis9kUFhaW7dCCa7k8OEMLPC+71Vg++OADnzoGBFkAcIP27dtzaVrg/8vppcS5PLjvSF2N5WppV2OpWrWqFyrLGEMLAAAAIMl6q7HQI4s8zZWZs3FxcRn+nBm+2gIA5FVWW42FIIs861pmzqaO18puH2bNAgDyKiutxsLQAuRphE0AAHLOKqux0COLPMvVmbNSzhblZmgBACCvs8pqLARZ5Gk5nTkLAACusMJqLAwtAAAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAluTzQfbs2bMaP368WrZsqQYNGujhhx/Wjh07vF0WAMAi1q1bp+bNm2vdunXeLgWwFCucOz4fZIcPH65ffvlFb7zxhj7//HPVrl1b/fv3V3R0tLdLAwD4uPj4eI0bN05HjhzRuHHjXLpkNQDrnDs+HWT//vtvbd68WRMmTJDdbtftt9+uF154QWXKlNGXX37p7fIAAD5u3rx5io2NlSTFxsZq/vz5Xq4IsAarnDsFvF1AVooXL6533nlHderUcWyz2WwyxujcuXNerAwA4OsOHDig+fPnyxgjSTLGaN68eerWrZsqVark3eIsyBjj1l65uLi4DH92l4CAANlsNre3mx9Y6dzx6SB78803q1WrVk7bvv76ax08eFAtWrTI9H7t2rXL9LZjx46pbNmybqsRAOB7jDEaP36844346u0ffPABISeH4uPjVatWLY+0bbfb3d5mVFSUAgMD3d5uXme1c8enhxZcLTIyUmPHjlW7du3Utm1bb5cDAPBR0dHR2rBhg5KTk522Jycna8OGDcyzADJhtXPHp3tk01q3bp2ee+451a9fX2+88UaW+0ZERGR6W1a9tQCAvKFKlSpq2bKlNm/e7PSG7O/vrxYtWqhKlSperM76lnXuoiL+uY8Qqb1+7urhu5ScpJ5rVrmlrfzKaueOJYLsRx99pFdeeUUdOnTQjBkzVKhQIW+XBADwYTabTS+99JLat2+f4XZf+mrUior4F1BAAUtECOSQ1c4dnx9a8Mknn+jll19Wr1699NZbbxFiAQAuqVSpkoYMGeJ447XZbBo6dKgqVqzo5coA32alc8eng2xMTIymTJmiDh06aNCgQTp16pROnjypkydP6sKFC94uDwDg44YOHaoyZcpIkoKCgjRkyBAvVwRYg1XOHZ8Ost9++60SExP13XffqUWLFk7/XnnlFW+XBwDwcQEBAZo8ebLKlSunl19+WQEBAd4uCbAEq5w7Pj3AZfDgwRo8eLC3ywAAWFj79u3TjfcDkD0rnDs+3SMLAAAAZIYgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgC1jEunXr1Lx5c61bt87bpQA+wRijuLi4bP/9+++/+vfff13a1xjj7V8LQA4U8HYBALIXHx+vcePG6fjx4xo3bpyaN2+ugIAAb5cFeI0xRj169FBkZKRb27Xb7QoLC5PNZnNruwA8gx5ZOKHXzzfNmzdPsbGxkqTY2FjNnz/fyxXlH/T6+S7CJgB6ZOFAr59vOnDggObPn+8IP8YYzZs3T926dVOlSpW8W1weR6+f77LZbAoLC1N8fHyW+8XFxclut0uSduzYocDAwCz3DwgI4LgAFkKPLBzo9fM9xhiNHz8+XQ9eZtvhfoQa32Wz2RQYGJjtv1Su7MvxBqyFHllIotfPV0VHR2vDhg3pticnJ2vDhg2Kjo5W1apVvVBZ/kCvHwD4NnpkQa+fD6tSpYpatmwpf39/p+3+/v5q1aqVqlSp4qXK8g96/QDAdxFk4ej1S05OdtqettcP3mGz2fTSSy+lCz6ZbQcAID8hyIJePx9XqVIlDRkyxBFabTabhg4dqooVK3q5MgAAvIsxsnD07rVv3z7D7fT6ed/QoUMVFham48ePKygoSEOGDPF2SQAAizLGZDv2P3U/ybVJr94a+0+QhaT/6/WbM2eOjDH0+vmYgIAATZ48WRMmTNCkSZNYFg0AcE3y2rKCDC2Aw9ChQ1WmTBlJotfPB7Vv316bN29O13MOAEBO5KVvWumRhQO9fgAA5G15bVlBgiyctG/fnh4/AADysNRlBV119TKDvoShBQAAALAkgiwAAAAsiSALAAAASyLIAgAAwJKY7AUAAGBhrl7gICfi4uIy/Nkd3LnCAUEWsIh169Y5lkZjZQnkVe5+Q7bKmzGQG/Hx8apVq5bH2k9dhstdoqKi3LYKAkEWsID4+HiNGzdOx48f17hx49S8eXPW+UWe5Mk3ZF9+MwZwbQiygAXMmzdPsbGxkqTY2FjNnz9fw4cP93JVAPKrS0lJ3i4hQ75a1/W0rHMXFfF3T7wzxkhyz5XALiUnqeeaVblu52oEWcDHHThwQPPnz3e8oBhjNG/ePHXr1k2VKlXybnEWlJ/HklmNu96QrfBmbAWpz6Mk9fza95+DtPXmJ0X8CyigQP6Jd/nnNwUsyBij8ePHp3tBTt3+wQcf5NuQc63y81gyq8lvb8gAco5XCMCHRUdHa8OGDem2Jycna8OGDYqOjlbVqlW9UBmA/Cbth+YPOnb2ud5y6UqPed+1a9zaJnwbQRbwYVWqVFHLli21efNmJScnO7b7+/urRYsWqlKlihers778NpYMcJfUsAjf46vjhD1VF0EW8GE2m00vvfRSuuW2UrfT45A7fHUNIC/Iz+OXeQUHfFylSpU0ZMgQzZkzR8YY2Ww2DR06VBUrVvR2aQDykYCAAEVFRbmtvbi4OMeY8h07drh9LDhLFOYPBFnAy1yZRf/YY49p2bJlOnHihEqXLq2+fftmOzs+P892B+B+NpvNYxMPAwMD8+2kRndI+1q/rFMXFfHBb5ouJSU5eovd+d7ke78pPMLVJYdyMs6PoJR7xhj16NFDkZGRLt8nNjZWDRs2zHY/u92usLAwjhEA5CNFCuSvIVP55zfNx64lLLmCoOQePH8AAHe5lOy+SVXunsTqCQTZfIKw5JtsNpvCwsKy7S2/lrFk9JgD8BZXvgW8lguJ8LqWvfy24glBNh/wVFjiBcU9cjrujLFk7pPflqkBrodr+RbQ1QuJ8E0grkaQzScIS8AV+XmZGqvxxUDvizX5IoLm9eXuFSUkz64q4c4VJQiyAACfYaUPGvn5Q0ZWXP0WUMr5GEy+CcyYJ1eUkHy7c4sgC3iAq6tEuOpaxpLlRH56c8jPy9QA14ungxWQyvdeweEyK4Wl/BSUJCk+Pl61atXySNuujiXLiaioqPz5puPGP0m3XjM+/5wq6aR9/j7o2NktlxB298zr1Muz5qfXNOQtnpiM5633eYKshVkpLOXboASflt9m91pNamAE4D6emoznrYl4BFnAw3yxV0ly7lkCAOQfeenbBIJsHuGLYSk/B6W0k0Cs8Bzkp0kr+Xl2rxW4+/hwbABnnpqMx9CCLKSkpGjOnDkKCwvT+fPn1bBhQ02YMEEVK1b0dmleZaWwlJ+CEnxbfp7dawWePD4cG+CKvDQZz8/bBbhi3rx5Wrp0qSZPnqxly5bJZrNpwIABSkhI8HZpQIas9rWN1eoFAECyQI9sQkKC3n//fY0cOVKtWrWSJL355pu688479d133+mee+7xcoXeY6XwYaVaAVhDXpp5DeDa+HyQ3b17t/799181bdrUse3mm29WrVq1tH379nwdZOG7XB3nZ4xR7969tWvXLrfX0KBBAy1ZssTlsU3IGGHJN+W1mdcAro3PB9njx49LksqWLeu0vXTp0jp27FiG92nXrl2m7R07dixdW1bl7bBEUMqcq+OPjDHy9/f3SA1+fn4KDAzkDTkXCEu+jecPgM8H2dSekEKFCjltL1y4sM6dO+eNknyGt8MSQSn3uJSj7+M59E15beY1gGvj80G2SJEikq6MlU39WZIuX76caS9fREREpu1l1VubV/GC79vy0uzRvIZzx7dx7gDw+SCbOgzgxIkTqlChgmP7iRMnVKNGDW+VZTm84APXhnMHAHyXzy+/VaNGDd14443aunWrY9v58+cVFRXlkWvOAwAAwBp8vke2UKFC6t27t2bMmKESJUqoXLlymj59uoKCgtShQwdvlwcAAAAv8fkgK0lPPfWUkpKSNG7cOF26dEmNGjXSwoUL000AAwAAQP5hiSDr7++vkSNHauTIkd4uBQAAAD7C58fIAgAAABkhyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsiyAIAAMCSCLIAAACwJIIsAAAALIkgCwAAAEsq4O0CrrcTJ04oOTlZ7dq183YpAAAAyMCxY8fk7++f7X75rke2cOHCKlAgb+T3Y8eO6dixY94uA5ng+Pgujo3v4tj4No6P78prx6ZAgQIqXLhwtvvZjDHmOtQDD0jtVY6IiPByJcgIx8d3cWx8F8fGt3F8fFd+PTb5rkcWAAAAeQNBFgAAAJZEkAUAAIAlEWQBAABgSQRZAAAAWBJBFgAAAJbE8lsAAACwJHpkAQAAYEkEWQAAAFgSQRYAAACWRJAFAACAJRFkM2GM0YoVK9SnTx81bdpUderUUfv27fXSSy8pNjbWsd/s2bNVvXr1LNu6fPmyJk2apGbNmik0NFRPPfWUTp06leV9Ll68qAkTJqhp06Zq2LChBg8erEOHDjnts3jxYnXo0EEhISHq2rWrfvjhh2v/hXNg9uzZatu2rdO2hIQENWjQwOm58YRrOS5jxoxJV+/Vzpw5oxEjRqhRo0Zq1KiRXnzxRcXFxWV5nxMnTmj48OGy2+1q0qSJRowYodOnTztuT0xM1JtvvqnWrVsrNDRUjzzyiHbu3JmL3951Y8aMUZ8+fdLVGxoaqoSEBLc+ljvPlYMHD2rIkCGy2+2y2+169tlndfz4cZfquHTpku677z6tWLEi031Onz6tFi1aaPbs2U7bd+7cqT59+qhhw4a688479cILL+js2bMuPW5uVa9ePV3NCxYs0Isvvpirdj11rvz+++/q27evQkND1bRpU40fP17nz593qabU53/r1q2Z7hMTE6PQ0NB0z8n333+v7t27KzQ0VG3bttWrr76qS5cuufS4uXH48GFVr149Xc3jx4/X22+/nau23XnupPXCCy9ozJgxWT5u//79071GXC0xMVF16tRR9erVnf69+eabjn0OHDiggQMHym63q0WLFho3bpzLfw+55Ylzx1PnTaqUlBT1798/3WtQWgkJCbrvvvuyPIaSdOTIkXTHpnr16goLC3Ps89tvv6l3794KDQ1Vq1at9Nprr7n9PSAjnjxvrkaQzUBycrKGDBmiadOmqU2bNlqyZInWrl2rF198UX/88Ye6d++uf/75R5Jks9mybW/ixInavHmzZs+erQ8++ECHDh3S008/neV9nnzySW3ZskWzZ8/Wxx9/rHPnzmnIkCFKSUmRJC1fvlwzZ87UiBEjtHr1arVu3VrDhg3Tn3/+mfsn4BpERkaqfPnyKlOmjMce41qPiyvH6KmnntKhQ4e0ePFizZo1S5s3b9akSZMy3T8hIUGPP/64Dh06pEWLFmnBggWKiorS6NGjHfvMnz9fy5cv1+TJkxUeHq7bb79dAwYM8HjYz8z69evVtGlTFSpUyG1tuvNcuXz5sh577DFJ0qeffqolS5bo5MmTGjRokLJbXOXs2bMaNGiQ9u7dm+V+L774ok6ePOm0LSYmRv3791eNGjUUFhamN998U7/++queeuqpbH57z1m/fr1atmx5zff31Lly4sQJ9evXTxUqVNDKlSs1b9487dy50+nvPjNHjx7VY489lu75TysxMVHPPfdcug+RO3bs0LBhw3TXXXcpPDxcEydO1Ndff53lOeppGzZs8Moxyq7NV199VZ9//nmW+33wwQfatGlTtu3t379fiYmJ+uKLL7Rp0ybHv4EDB0q6crwGDhyoggULatmyZXrrrbe0fft2jRs3zqV6PSE3544n32OkKx+2R44cme1z/9prr2X7WiZJe/bsUeHChbVx40an43PfffdJuvLB8YknntDtt9+u8PBwvfzyy1q5cqXTB5HrLbfnTUYIshlYtGiRNm7cqEWLFunxxx9XtWrVdOutt6pVq1ZavHixChYsqPfff1+SdMstt6hUqVKZthUbG6vw8HCNGzdOdrtd9erV0xtvvKHt27fr559/zvA+W7dudYTYRo0aqUaNGnr55Zf177//6sCBA5KkiIgItWjRQnfffbduu+02Pf3007rhhhu0ZcsWdz8dLlm/fr1atWrl0ce41uNyyy23qHTp0pm2u2vXLm3btk1Tp05V7dq11axZM7300kv64osvMg2dX375pY4cOaL58+erbt26CgkJ0dixYxUTE6OLFy9KunKM7r33XrVo0UIVK1bUmDFjdPHixUyPu6dt2LDB7cfInefK0aNHVbduXb3yyiuqVq2aatasqccee0y7d+/WmTNnMr1fRESE7r//fl24cCHLWpctW6aYmJh0NYSHh6t06dIaO3asbr/9dtntdk2YMEFbt25N9y3I9XD+/Hn98ccfatas2TW34alz5ciRI7rzzjs1YcIEVapUSQ0aNNCDDz6Y7etOWFiYHnjgARUoUCDL/WbPnq0bbrgh3falS5eqadOmGjhwoCpWrKiWLVvq2Wef1apVq65L79LV9uzZo6SkJNWsWfOa23DnuSNJ0dHRevjhhxUeHq5bb701y9rnzp2rkJCQbGvcu3evbrrpJtWoUUOlSpVy/Es9Rvv27dPff/+tp556SlWqVJHdblevXr20ceNG158IN8rtueOp80a68q1P165d9csvv+jmm2/OdL+NGzfq66+/VrVq1bKtd+/evapcubJKly7tdHyKFCnieMyzZ89q1KhRjvOmS5cuLn2I8QR3nDcZIchexRijjz/+WF26dFHt2rXT3R4QEKCPPvpIzzzzjCSpatWqqlq1aqbtRUZGSpKaNGni2Fa5cmWVKVNG27dvz/A+GzduVHBwsNNXSVWrVtUPP/yg22+/XZJUrFgxbd++Xbt375YxRmvWrNGFCxdUr169DNtM7eafN2+emjdvrrZt2+r8+fP666+/NHToUDVp0kR16tRRhw4d9MEHHzjdd9myZerQoYPq1aunoUOH6ty5c+naTxtkDxw4oP79+6thw4YKDQ1V//79tWfPnkyfI1fk5rhUqVJFVapUybTtHTt2qFSpUk77NG7cWDabzXH8rrZx40Y1bdpUt9xyi2PbnXfeqXXr1unGG2+UdOUY/fDDDzp8+LCSk5O1bNkyFSpUKNOTeOvWrapevbreffddNWnSRF27dlVycrIiIyPVr18/NWzYUHXq1NG9996rL7/80um5mTdvnlq2bKmQkBC98MILunz5slPbiYmJ+t///uc4Rr/++qseeeQRhYaGqlGjRnryySd19OjRTJ+jjLj7XKlcubJmzpypEiVKSLryN/vJJ5+odu3aKl68eKb3+/HHH9WnTx8tXbo0031iYmI0Y8YMTZ8+PV2PdJcuXfTqq69m2KuS2fCCMWPGaNiwYXr88cfVoEEDLViwQMYYvffee+rUqZPq1Kmjhg0batCgQU5h+Pjx4xoyZIhCQ0PVunVrffXVV+na3rRpk0JCQhx/RwsXLlT79u1Vp04dtW3bVnPnzs2yh9qT50poaKjeeOMNRyDdt2+fVq5cqebNm2d6H0n64YcfNHLkSM2cOTPTfbZv365ly5bp1VdfTXfb448/rlGjRqXbnpSU5PjgeLU+ffpo7NixevDBB2W32xUeHq6EhAS9/vrrjuezSZMmGj58uNMHpb179+rRRx9VSEiI7rrrLv3000/p2k7t9bPZbEpOTtb06dPVqlUr1alTR3fffbc+/fTTLJ8Pd587krRt2zbVrFlTX375pcqXL5/hPpcvX9Zzzz2np556SpUrV86yPelK8MjqcYsWLSqbzaawsDAlJCTo9OnT+vbbb7MMyb567njyvJGuvGd06NBB4eHhuummmzLc5/Tp03r++ef18ssvZ/malyq741OsWDFJV77hSk5O1uHDh7V+/XrVr18/0/v48nmTKQMnBw8eNMHBwWbVqlVuae/99983zZo1S7e9e/fuZtKkSRne57///a8ZNmyY+fjjj03nzp1NixYtzNNPP22OHz/u2OfYsWOmZ8+eJjg42NSsWdNUr17dfPLJJ5nWcejQIRMcHGzuuusu89dff5lff/3VxMXFmebNm5sRI0aYffv2mQMHDpjXX3/dBAcHm6ioKGOMMV9++aWpVauW+eijj8z+/fvNggULTI0aNUybNm2c2rbb7SYxMdEYY0zXrl3NmDFjTExMjPnrr7/ME088Ydq3b39Nz18qdx+XtF5++WXTo0ePdNubNm1q3nvvvQzv88ADD5jJkyebOXPmmPbt25vWrVubcePGmXPnzjn22bNnj+ncubPjGNWuXdusW7cu0zp++uknExwcbB555BETExNjoqKizPHjx029evXMtGnTzIEDB8y+ffvMmDFjTJ06dczJkyeNMca8/fbbJjQ01KxevdpER0ebKVOmmODgYNO7d29H21u2bDH33nuvMcaY5ORk07RpU/PGG2+YgwcPmt9//91069bN9O3bN0fPmyePSb9+/UxwcLBp1KiR42/RFcHBwWb58uVO2xISEkzXrl3N/PnzjTHGtGnTxsyaNSvLdgYOHGiaN29uEhISMrx99OjRJjg42Lz77rtm//795ujRo2bRokXGbrebiIgIc/jwYfPTTz+ZDh06mKFDhxpjjElMTDT33HOP6dmzp/n999/Nzp07zf3335+u5lGjRpmFCxcaY4yJiIgwdrvdbNq0yRw5csR89dVXpnbt2iY8PDzT2j15XNLq2LGjCQ4ONm3atDFHjhxx6T6pr0M//fST0/Zz586Z1q1bO2rO6DimdfnyZXP//febrl27ZrpP7969TfXq1c2qVavM3r17zenTp83LL79sWrdubX766Sdz+PBh88MPP5jGjRubV155xRhjzPnz502zZs3MkCFDzN69e82mTZtMmzZt0tXcq1cv88033xhjjPnwww9N27ZtTWRkpDl8+LBZsmSJCQ4ONtu3b8+0Nk8fo969e5vRo0en2/7yyy+bxx9/3KSkpJjRo0c7vUZkZMCAAaZ79+7m8ccfN82aNTNdu3ZN97f3wQcfmJCQEFOjRg0THBxs7rnnHnP69OlM2/TVc+d6nTfGZP4aNHjwYDNu3DhjTObHMK3OnTubvn37mocfftg0a9bMPPTQQ2bDhg1O+8yYMcPUrl3b1KxZ0/G+cOnSpUzb9OXzJjNZf8+TD6WOf0ntFUo1ePBgp0HLt956a4afCK8WHx+f4ZjEwoULp+s1S3Xx4kX98ccfOnPmjGMM2IwZM/Too49q1apVKly4sA4ePChjjF577TVVq1ZNa9eu1ZQpU1ShQoUse0ceeeQRxye406dP69FHH9Ujjzzi+AQ7bNgwLViwQHv27FHNmjX14YcfqnPnzurVq5ckaeDAgfr555+1e/duR5vr169X8+bNHb00Bw8eVPPmzVW+fHkVKFBAU6ZM0f79+5WSkiI/v2v7EsDdxyWtaz1G4eHhatasmV5//XWdO3dOU6dO1dChQ7VkyRLZbDZFR0fr5ptv1ty5c1WmTBmFhYVp9OjR+uijj1SjRo1M63n88cdVqVIlSdKhQ4c0bNgw9e/f3/HcDRo0SCtWrNCBAwdUsmRJLVmyRI8++qjuvfdeSdLzzz+fboB92nFJFy5c0JkzZ1S6dGmVL19eNptNb731VrYTEK/myWMycuRIPf3005o/f74ee+wxhYeHq2zZsjlqI9WsWbNUuHBhDRgwwKX9p02bpvXr12vWrFkqWLBgpvsVLVpUTzzxhOP/FSpU0LRp0xyTPsqVK6dOnTo5fvctW7bor7/+0nfffacKFSpIkqZOnaoHHnjA0YYxxmkM4sGDB1W4cGGVL19et956q2699VaVLl06y6+OPXlc0poxY4YuXbrkeG364osvMhwW4IqJEycqJCTEMbYvK0lJSRo1apT27dunjz/+OMt9a9as6dRm3bp11bFjRzVu3FjSlWPUokULxzdGX331leLj4/Xqq6/qpptuUrVq1TR27Fj997//dbRx8eJF/fbbb47X2YMHDyowMFC33XabSpUqpd69e+v222/Pssfzeh2jtDZs2KDVq1dr1apVLo/p/Ouvv1SgQAE99dRTKlWqlH788Uc9//zzSkxMVI8ePZSQkKC//vpLHTt2VK9evXTmzBm99tprevbZZ/X+++9n+nrvi+eON45JWkuXLlV0dLRef/11l/ZPSEjQgQMHFBAQoFGjRikwMFCrVq3SgAEDtGjRIjVr1kznz5/XgQMH1KtXL3Xp0kWHDh3S1KlTNXHiRE2dOjXTtn31vMkMQfYqqd35V3+lOGnSJMcM2SVLluj7779Pd98nnnjC6avoSZMmqUiRIhmO4bp8+bICAgIyrKFgwYK6fPmy5s6dq6JFi0qS5syZozvvvFPff/+9WrZsqSeffFLPP/+87r//fklSrVq1dOTIEU2fPj3LIFuxYkXHzyVKlNAjjzyiNWvWaPfu3fr7778dk8VSJ5Xt3btX99xzj1MboaGhTkF2w4YNuuuuuxz/f/bZZzVlyhR9+umnatq0qe6880516tTpmkOslLvjktY999zj9BX6u+++m+UxCgwMzLCdggULKjAwUK+//roj7BQtWlQPPvigfvvtN5UsWVIjR47U4sWLZbfbJV15Mdi3b59mz56tuXPnZlpjaoiVpNtuu03du3fXRx99pH379unAgQOOY5ScnKwzZ87o5MmTqlu3rlMbISEhio6Odvx/w4YNjpm8qW8iL7/8subMmaM77rhDLVu2dDqGrnD3udKlSxfH/1OHX6Su+rB8+XINGzYsR/VJV75u/fTTT7Vy5Ur5+/tnuW9iYqLGjx+vlStXasKECerYsWOW+6c9lySpbdu2+uWXXzRr1iz9/fffio6O1l9//eWYALl3714VLVrU8Uac+numfR34/fffVbhwYcfXlF26dNHy5cvVsWNHVa9eXc2bN1eHDh2yDLKePFdS/5YlOf7mZs+erVatWum7775zChauCg8P144dO7R69eps97148aKeeeYZbd26VbNmzcryK1Ip/TG6//77tWXLFr3xxhs6cOCAoqOjtX//fsfvtXfvXlWqVMnpq9/Q0FCnNjZt2qR69eo5Pvz36tVL69atU8uWLVWnTh01b95cnTp1UsmSJTOty5PnTkZOnz6tsWPHauLEiTmakPvNN98oJSXF8Tdas2ZNHTt2TAsXLlSPHj20aNEibdu2TWvWrHGcX5UqVVLHjh31/fffq3379hm264vnzvU6bzKyf/9+TZ8+XQsXLsz0PedqhQoV0vbt21WgQAFHR0ydOnUUHR2thQsXqlmzZpoxY4bOnz/vWCGhdu3aKlq0qB577DH17ds30w4VXz1vMkOQvUrqp4Nt27Y5Bbi0J39quLzaK6+84rQcTMmSJbVhwwadPXtWCQkJTr1+J06cUFBQUIbtBAUFqUyZMk6Pc8stt6hYsWI6fPiwoqOjdfbs2QzDy3fffZfl75c6CFy68gn0P//5j4oXL6527dqpWbNmqlu3broJQeaqMUVpe6kuX76sbdu26ZVXXnFs69Wrl+6++26tX7/e8cc/e/ZshYeHO40pzYncHJe03nnnHSUlJTndf9euXVq3bp3TfgkJCTp79mymL/pBQUFKSUlxei5SB+cfPnxYR44cUWJiYrpjVL9+fW3YsCHLGgsXLuz4OXUCR61atdS8eXO1a9dOxYsX14MPPuh0n6uPUdpJNUePHtXx48fVoEEDx7bnnntOjzzyiOMYTZw4UQsWLFB4eLjLqxq4+1w5cuSIfv/9d6dAHRAQoPLly+vEiRMu1XS1lStXKi4uzumNPj4+XgsWLND777+vXbt2SboSkIYNG6YdO3bo9ddfT/fhLSNpzyXpyhvW7Nmz1a1bNzVu3Fh9+vRRRESEU+/N1cdJcj5WV8+4LlGihL744gvt2rVLmzdv1qZNm/T+++/rySefzDTYe/JciY6O1uHDh51eI0qXLq2iRYte82ocy5cv16lTp9S6dWun7RMmTNDChQsdz9+JEyc0YMAAHT58WO+++66aNm2abdtXH6OJEydqzZo1euCBB9S6dWsNGTJECxcudKo9q3NJSn+MKlWqpLVr12rbtm3avHmzIiIi9Pbbb2vq1Knq2rVrhnW5+9zJzvr163Xy5EmNHTtWY8eOlXTlNS4lJUWhoaGZhq20r0Wpqlev7vjQERkZqVq1ajl9SKxYsaKKFy/umJicEV88dzx53mRnzZo1+vfff9WvXz/HtkuXLmnnzp369ttv9dVXX2UYwDMKvcHBwY7JXJGRkWrTpo3T7akf/mJiYjINsr563mSGyV5X8ff316OPPqrw8HCnXse0jh07luH2MmXKqGLFio5/N954oxo2bKiUlBSnT9D79+9XbGxspp/S7Ha7jh496vTmfeLECZ05c0YVK1Z0fMV69QSqPXv2pPsklZXVq1fr7NmzWrp0qYYOHaoOHTo4JnKl/lHWrFkz3YSn3377zfHz1q1bVblyZUdA/eeff/TSSy8pMTFR3bp10/Tp07Vq1SqdPHlS27Ztc7m2q+XmuKRVrlw5p2NUpEgRNWrUSMePH9fff//t9HtJcgp/adntdu3evdvpDSV1uZSsjtHevXtzdIw+/fRTlSxZUosXL9aAAQPUqlUrx1dgxhiVKFFCZcuWTXeMfv/9d8fP69ev1x133OEI3fv379eECRNUsmRJPfzww5o1a5bee+89RUdHZ/rcZsTd58qff/6pp556SgcPHnTsd/78ecXExGQ7kSIzzz33nL7++muFh4c7/pUuXVoPPfSQwsPDJV15Qx80aJB+++03vffeey6F2IzMnz9fw4YN08SJE9WzZ0+FhITowIEDjnOpVq1ajgmWqWJiYpxWW7h6ZYkvvvhCn376qRo2bKinnnpKn332mR588EGtWbMm0zo8ea5s3LhRTz/9tNMEq4MHD+rMmTPXfIxmzJihNWvWOB0j6cqSeO+8844k6dy5c+rbt69Onz6tTz75xKUQe7UzZ87o008/1cSJEzV27Fh169ZNNWvW1P79+51e72JiYpzWg077emeM0caNG52O0Ycffqi1a9eqefPmGjVqlFavXq1mzZp57BhldO5kp0OHDlq7dq3Tc9y2bVvVqVNH4eHhqlOnTrr7nD17Vna7XV988YXT9t9++83xoT0oKEh//fWXU4iJjY3V2bNnnb5Zyo4vnDuePG+y07t3b3377bdOxyd1glrqa9bVdu/erdDQUO3YscNp+++//+4YPhgUFJThe5Akl4+PL503maFHNgNPPPGEoqKi9Mgjj2jgwIFq3bq1brzxRu3du1cfffSRNm/erO7du7vUVpkyZXTPPfdo3LhxmjJligICAjRhwgQ1btzYMbMzISFB586dU9GiRVWoUCF16tRJ77zzjp5++mmNHTtW/v7+mjJliipXrqzWrVurUKFCuvfeezVlyhQVLlxYwcHB+uGHH7R8+XLNmDHD5d8zKChI8fHx+vrrr2W327V//37HuJnUr9oHDhyoIUOG6L333lP79u21ceNGffvtt44T6+oXj2LFiunHH3/UwYMHNWLECN144436/PPPVbBgwQxfLHPCncclrfr166tBgwZ69tlnNXHiRMXFxWnChAl64IEHHJ+mL126pAsXLqhEiRLy9/fXQw89pI8//lgjRozQ008/rQsXLmjixIlq0qSJateurZSUFNntdo0ePVoTJkxQUFCQwsPDtWXLFn3yyScu1xYUFKTjx49r/fr1qlq1qv744w9NnjxZ0v8dowEDBujVV191LB31xRdf6Ndff1XDhg0lXQmyab/iK1asmL788ktdunRJAwcOlJ+fn5YvX66iRYs6VsVwlTuPScuWLVW9enWNGjVKL774oowxmj59uooXL+5o4+pzJTslS5ZM12NVoEABFS1a1PGBYsGCBYqMjNTrr7+uKlWqOK1z6urjSFLZsmW1efNmtW3bVn5+fvriiy+0du1ax4e8Jk2aqH79+ho1apQmTJggf39/TZ482THk5vTp09q7d69TSLt8+bJeffVV3XDDDbLb7Tp+/Li2bdumRo0aZVmLp86V+++/XwsXLtTIkSM1fPhwnTt3TpMnT1a9evUcPT9XnyvZyazHqmTJkipXrpykK+MhDx06pPfee08lSpRwOkauPs5NN92km266SREREapdu7YuXbqkjz76SH/88Yejl+qee+7R/PnzNWLECI0ePVrnz5/XlClTHG1ERUWpQIECCg4Odmw7deqU5s6dqyJFiqhGjRqKjo5WVFSU+vbtm2U9njpGGbnxxhvTBd4bbrhBRYoUcfpgnfq1erFixVSsWDHdcccdeuONN1SiRAnddtttWrt2rVatWqUFCxZIuhLAVq5cqRdffFH9+vXThQsXNHXqVFWvXj1HS/35yrlzPY9JWqnPd1pFihTRDTfc4HR8Tp8+rYIFC+qmm25ScHCwqlWrpkmTJmnChAkqXry4PvvsM/3888+OdYT79eunJ554Qm+99Za6deumI0eOaNKkSWrVqpXLS2D52nmToRxPD8tH1qxZY/r372+aNWtmateubVq1amVGjBhhtm3blqN2/v33X/PCCy8Yu91u7Ha7GT58uNOsztTZ6mln9504ccIMHz7cNGjQwISGhpphw4Y5rVoQHx9v3njjDdO+fXsTEhJiunbt6pgNmJGMZgunpKSY6dOnmzvuuMPUq1fPdOrUybz//vumV69eZsyYMY79vvrqK3P33XebOnXqmN69e5vXX3/dsWpBhw4dzK5du5wea9++fWbAgAGmcePGpm7duuY///mP2bx5c46es6y467ik9c8//5gnn3zShISEmCZNmpgJEyY4zexcvny5CQ4ONocOHXJsi4mJMQMHDjT169c3jRo1Ms8//7w5f/684/azZ8+aiRMnmtatW5vQ0FDTs2dPs3Xr1kxrSP07SPsYly9fNs8//7xp3LixCQkJMQ888IBZuXKladOmjZk9e7Zjvw8//NC0adPG1K1b1wwdOtS8+OKLpnfv3uby5csmJCTExMbGOj3Wzp07Ta9evUyDBg1M/fr1Td++fc3vv/9+zc+fu45JbGysGT58uGnSpIkJDQ01Tz75pDl69Kjj9ozOlbSym+1uTPoZw6mz7zP6l9njZDTj+/fffzf/+c9/TL169UyzZs3M4MGDzaeffmqqV6/uOKanT582w4cPN6GhoaZZs2Zm8eLF5o477jDLly83X3zxhXn88cfTPdY777xjOnToYOrUqWOaN29uXnrpJRMXF5fl75jKE+fK/v37zcCBA03Dhg1N48aNzfPPP++0WkdG50qqzFYtuFra45icnGzq1q2b6THK6HGMyXjW96ZNm8y9995r6tata1q0aGFGjBhh3n77bRMSEmL+/fdfR40DBgwwISEhplWrVmbFihWOmufNm2defPFFpzYTEhLMa6+9Zlq1amVq165tWrdubd566y2TlJSU/ZNpPHOMXJnxntHfcO/evZ22/fvvv2batGmmVatWpk6dOub+++833333ndN9du3aZfr06WMaNmxoWrRoYcaMGWNOnTqVo8f1tXPHE8ckLVdWTsnoGLZp08Zp26lTp8zzzz9vmjdvburWrWt69uyZbtb/jz/+aB588EETEhJi2rRpY6ZMmeL4W3f1cX3xvEnLZkw2l8wBAAAAfBBjZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCURZAEAAGBJBFkAAABYEkEWAAAAlkSQBQAAgCX9P8FF1vzqC0dPAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>Frequency Sweep G' at 0.68 rad/s (Pa)</th>\n",
       "      <th>Frequency Sweep G'' at 0.68 rad/s (Pa)</th>\n",
       "      <th>Time Sweep G' average (Pa)</th>\n",
       "      <th>Time Sweep G'' average (Pa)</th>\n",
       "      <th>Frequency Sweep G' at 14.58 rad/s (Pa)</th>\n",
       "      <th>Frequency Sweep G'' at 14.58 rad/s (Pa)</th>\n",
       "      <th>Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)</th>\n",
       "      <th>Condition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.099382</td>\n",
       "      <td>0.096231</td>\n",
       "      <td>0.207591</td>\n",
       "      <td>0.152495</td>\n",
       "      <td>0.562104</td>\n",
       "      <td>0.233578</td>\n",
       "      <td>0.094778</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.162064</td>\n",
       "      <td>0.239388</td>\n",
       "      <td>0.432566</td>\n",
       "      <td>0.367460</td>\n",
       "      <td>0.842844</td>\n",
       "      <td>0.442773</td>\n",
       "      <td>0.209570</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.621684</td>\n",
       "      <td>0.575987</td>\n",
       "      <td>0.881257</td>\n",
       "      <td>0.673098</td>\n",
       "      <td>1.092634</td>\n",
       "      <td>0.713464</td>\n",
       "      <td>0.542465</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.297534</td>\n",
       "      <td>0.339800</td>\n",
       "      <td>0.559338</td>\n",
       "      <td>0.489490</td>\n",
       "      <td>0.671828</td>\n",
       "      <td>0.580777</td>\n",
       "      <td>0.297843</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.000688</td>\n",
       "      <td>0.008248</td>\n",
       "      <td>0.020377</td>\n",
       "      <td>0.039847</td>\n",
       "      <td>-0.033777</td>\n",
       "      <td>0.117268</td>\n",
       "      <td>0.000658</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999995</th>\n",
       "      <td>0.031694</td>\n",
       "      <td>0.079739</td>\n",
       "      <td>0.119927</td>\n",
       "      <td>0.198677</td>\n",
       "      <td>0.277235</td>\n",
       "      <td>0.342409</td>\n",
       "      <td>0.069414</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999996</th>\n",
       "      <td>0.274292</td>\n",
       "      <td>0.241739</td>\n",
       "      <td>0.401000</td>\n",
       "      <td>0.316981</td>\n",
       "      <td>0.476396</td>\n",
       "      <td>0.396224</td>\n",
       "      <td>0.226724</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999997</th>\n",
       "      <td>0.882128</td>\n",
       "      <td>0.711955</td>\n",
       "      <td>1.053992</td>\n",
       "      <td>0.783176</td>\n",
       "      <td>1.139733</td>\n",
       "      <td>0.829240</td>\n",
       "      <td>0.650164</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999998</th>\n",
       "      <td>1.263681</td>\n",
       "      <td>0.934258</td>\n",
       "      <td>1.327501</td>\n",
       "      <td>0.980105</td>\n",
       "      <td>1.264398</td>\n",
       "      <td>1.043896</td>\n",
       "      <td>0.835115</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999999</th>\n",
       "      <td>0.164244</td>\n",
       "      <td>0.180741</td>\n",
       "      <td>0.301407</td>\n",
       "      <td>0.280149</td>\n",
       "      <td>0.464487</td>\n",
       "      <td>0.379625</td>\n",
       "      <td>0.163333</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2000000 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Frequency Sweep G' at 0.68 rad/s (Pa)  \\\n",
       "0                                     0.099382   \n",
       "1                                     0.162064   \n",
       "2                                     0.621684   \n",
       "3                                     0.297534   \n",
       "4                                    -0.000688   \n",
       "...                                        ...   \n",
       "1999995                               0.031694   \n",
       "1999996                               0.274292   \n",
       "1999997                               0.882128   \n",
       "1999998                               1.263681   \n",
       "1999999                               0.164244   \n",
       "\n",
       "         Frequency Sweep G'' at 0.68 rad/s (Pa)  Time Sweep G' average (Pa)  \\\n",
       "0                                      0.096231                    0.207591   \n",
       "1                                      0.239388                    0.432566   \n",
       "2                                      0.575987                    0.881257   \n",
       "3                                      0.339800                    0.559338   \n",
       "4                                      0.008248                    0.020377   \n",
       "...                                         ...                         ...   \n",
       "1999995                                0.079739                    0.119927   \n",
       "1999996                                0.241739                    0.401000   \n",
       "1999997                                0.711955                    1.053992   \n",
       "1999998                                0.934258                    1.327501   \n",
       "1999999                                0.180741                    0.301407   \n",
       "\n",
       "         Time Sweep G'' average (Pa)  Frequency Sweep G' at 14.58 rad/s (Pa)  \\\n",
       "0                           0.152495                                0.562104   \n",
       "1                           0.367460                                0.842844   \n",
       "2                           0.673098                                1.092634   \n",
       "3                           0.489490                                0.671828   \n",
       "4                           0.039847                               -0.033777   \n",
       "...                              ...                                     ...   \n",
       "1999995                     0.198677                                0.277235   \n",
       "1999996                     0.316981                                0.476396   \n",
       "1999997                     0.783176                                1.139733   \n",
       "1999998                     0.980105                                1.264398   \n",
       "1999999                     0.280149                                0.464487   \n",
       "\n",
       "         Frequency Sweep G'' at 14.58 rad/s (Pa)  \\\n",
       "0                                       0.233578   \n",
       "1                                       0.442773   \n",
       "2                                       0.713464   \n",
       "3                                       0.580777   \n",
       "4                                       0.117268   \n",
       "...                                          ...   \n",
       "1999995                                 0.342409   \n",
       "1999996                                 0.396224   \n",
       "1999997                                 0.829240   \n",
       "1999998                                 1.043896   \n",
       "1999999                                 0.379625   \n",
       "\n",
       "         Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)  Condition  \n",
       "0                                               0.094778          1  \n",
       "1                                               0.209570          1  \n",
       "2                                               0.542465          0  \n",
       "3                                               0.297843          1  \n",
       "4                                               0.000658          1  \n",
       "...                                                  ...        ...  \n",
       "1999995                                         0.069414          1  \n",
       "1999996                                         0.226724          1  \n",
       "1999997                                         0.650164          0  \n",
       "1999998                                         0.835115          0  \n",
       "1999999                                         0.163333          1  \n",
       "\n",
       "[2000000 rows x 8 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Preprocess data by loading the primary data file \"Statistical_Analysis_Database.xlsx\"\n",
    "database = preprocess_data(primary_data_file=\"Data/Statistical_Analysis_Database.xlsx\")\n",
    "\n",
    "# Filter and clean dataframes based on certain conditions\n",
    "# Clean data\n",
    "clean_df = database[(database[\"Joint\"] == \"knee\") & (database[\"Blood\"] == \"no\") & (database[\"Tissue or Clot in Test\"] == \"no\")]\n",
    "clean_df = clean_df.drop(columns=[\"Joint\", \"Blood\", \"Clot\", \"Tissue\", \"Tissue or Clot in Test\", 'Storage in freezer', 'COVID 19 now', 'COVID 19', 'Vaccined COVID 19'])\n",
    "\n",
    "# Create separate dataframes for different conditions\n",
    "OA_df = database[(database[\"Joint\"] == \"knee\") & (database[\"Condition\"] == \"oa\") & (database[\"Blood\"] == \"no\") & (database[\"Tissue or Clot in Test\"] == \"no\")]\n",
    "OA_df = OA_df.drop(columns=[\"Joint\", \"Clot\", \"Tissue\", \"Tissue or Clot in Test\", 'COVID 19 now', 'COVID 19', 'Vaccined COVID 19'])\n",
    "healthy_df = database[(database[\"Joint\"] == \"knee\") & (database[\"Condition\"] == \"healthy\") & (database[\"Blood\"] == \"no\") & (database[\"Tissue or Clot in Test\"] == \"no\")]\n",
    "healthy_df = healthy_df.drop(columns=[\"Joint\", \"Clot\", \"Tissue\", \"Tissue or Clot in Test\", 'COVID 19 now', 'COVID 19', 'Vaccined COVID 19'])\n",
    "\n",
    "# Combine dataframes\n",
    "clean_df = pd.concat([healthy_df, OA_df], axis=0, ignore_index=True)\n",
    "\n",
    "# Further processing of dataframes...\n",
    "OA_blood_df = database[(database[\"Joint\"] == \"knee\") & (database[\"Condition\"] == \"oa\") & (database[\"Blood\"] == \"yes\") & (database[\"Tissue or Clot in Test\"] == \"no\")]\n",
    "OA_blood_df = OA_blood_df.drop(columns=[\"Joint\", \"Clot\", \"Tissue\", \"Tissue or Clot in Test\", 'COVID 19 now', 'COVID 19', 'Vaccined COVID 19'])\n",
    "OA_total_df = pd.concat([OA_blood_df, OA_df], axis=0, ignore_index=True)\n",
    "healthy_blood_df = database[(database[\"Joint\"] == \"knee\") & (database[\"Condition\"] == \"healthy\") & (database[\"Blood\"] == \"yes\") & (database[\"Tissue or Clot in Test\"] == \"no\")]\n",
    "healthy_blood_df = healthy_blood_df.drop(columns=[\"Joint\", \"Clot\", \"Tissue\", \"Tissue or Clot in Test\", 'COVID 19 now', 'COVID 19', 'Vaccined COVID 19'])\n",
    "healthy_total_df = pd.concat([healthy_blood_df, healthy_df], axis=0, ignore_index=True)\n",
    "total_df = pd.concat([healthy_total_df, OA_total_df], axis=0, ignore_index=True)\n",
    "OA_storage_df = OA_df.copy().reset_index(drop=True)\n",
    "OA_male_df = OA_df.copy().loc[database[\"Gender\"] == \"male\"].reset_index(drop=True)\n",
    "OA_female_df = OA_df.copy().loc[database[\"Gender\"] == \"female\"].reset_index(drop=True)\n",
    "\n",
    "# Apply logarithm to specific columns\n",
    "columns_to_log = [\n",
    "    \"Frequency Sweep G' at 0.68 rad/s (Pa)\",\n",
    "    \"Frequency Sweep G'' at 0.68 rad/s (Pa)\",\n",
    "    \"Time Sweep G' average (Pa)\",\n",
    "    \"Time Sweep G'' average (Pa)\",\n",
    "    \"Frequency Sweep G' at 14.58 rad/s (Pa)\",\n",
    "    \"Frequency Sweep G'' at 14.58 rad/s (Pa)\",\n",
    "    \"Flow Step Viscosity at 3.162 1/s (Pa.s)\",\n",
    "    \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\",\n",
    "    \"Cross Over Point (rad/s)\",\n",
    "    \"tan(delta) at 0.68 rad/s\",\n",
    "    \"tan(delta) at 3.142 rad/s\",\n",
    "    \"tan(delta) at 14.58 rad/s\"\n",
    "]\n",
    "OA_df_log = apply_logarithm(OA_df.copy(), columns_to_log)\n",
    "healthy_df_log = apply_logarithm(healthy_df.copy(), columns_to_log)\n",
    "clean_df_log = pd.concat([healthy_df_log, OA_df_log], axis=0, ignore_index=True)\n",
    "OA_total_df_log = apply_logarithm(OA_total_df.copy(), columns_to_log)\n",
    "healthy_total_df_log = apply_logarithm(healthy_total_df.copy(), columns_to_log)\n",
    "total_df_log = pd.concat([healthy_total_df_log, OA_total_df_log], axis=0, ignore_index=True)\n",
    "OA_blood_df_log = apply_logarithm(OA_blood_df.copy(), columns_to_log)\n",
    "OA_storage_df_log = apply_logarithm(OA_storage_df.copy(), columns_to_log)    \n",
    "OA_male_df_log = apply_logarithm(OA_male_df.copy(), columns_to_log)\n",
    "OA_female_df_log = apply_logarithm(OA_female_df.copy(), columns_to_log)\n",
    "\n",
    "# Prepare dataframes for plotting and legend replacement\n",
    "# Define the value_vars and replacements for each category\n",
    "value_vars1 = [\"Frequency Sweep G' at 0.68 rad/s (Pa)\", \"Frequency Sweep G'' at 0.68 rad/s (Pa)\",\n",
    "               \"Time Sweep G' average (Pa)\", \"Time Sweep G'' average (Pa)\",\n",
    "               \"Frequency Sweep G' at 14.58 rad/s (Pa)\", \"Frequency Sweep G'' at 14.58 rad/s (Pa)\"]\n",
    "replacements1 = {\n",
    "    \"Frequency Sweep G' at 0.68 rad/s (Pa)\": \"G' at 0.68 rad/s\",\n",
    "    \"Frequency Sweep G'' at 0.68 rad/s (Pa)\": \"G'' at 0.68 rad/s\",\n",
    "    \"Frequency Sweep G' at 14.58 rad/s (Pa)\": \"G'' at 14.58 rad/s\",\n",
    "    \"Frequency Sweep G'' at 14.58 rad/s (Pa)\": \"G' at 14.58 rad/s\",\n",
    "    \"Time Sweep G' average (Pa)\": \"G' at 3.142 rad/s\",\n",
    "    \"Time Sweep G'' average (Pa)\": \"G'' at 3.142 rad/s\"\n",
    "}\n",
    "value_vars2 = [\"Flow Step Viscosity at 3.162 1/s (Pa.s)\", \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\"]\n",
    "replacements2 = {\n",
    "    \"Flow Step Viscosity at 3.162 1/s (Pa.s)\": \"n\",\n",
    "    \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\": \"|n*|\"\n",
    "}\n",
    "value_vars3 = [\"Cross Over Point (rad/s)\"]\n",
    "replacements3 = {\n",
    "    \"Flow Step Viscosity at 3.162 1/s (Pa.s)\": \"n\",\n",
    "    \"Cross Over Point (rad/s)\": \"Cross Over Point\"\n",
    "}\n",
    "value_vars4 = [\"tan(delta) at 0.68 rad/s\", \"tan(delta) at 3.142 rad/s\", \"tan(delta) at 14.58 rad/s\"]\n",
    "replacements4 = {\n",
    "    \"tan(delta) at 0.68 rad/s\": \"tan(delta) at 0.68 rad/s\",\n",
    "    \"tan(delta) at 3.142 rad/s\": \"tan(delta) at 3.142 rad/s\",\n",
    "    \"tan(delta) at 14.58 rad/s\": \"tan(delta) at 14.58 rad/s\"\n",
    "}\n",
    "\n",
    "# Prepare DataFrames for plotting and legend replacement\n",
    "clean_osc = prepare_dataframe(clean_df, \"Condition\", value_vars1, replacements1)\n",
    "clean_n = prepare_dataframe(clean_df, \"Condition\", value_vars2, replacements2)                                 \n",
    "clean_cop = prepare_dataframe(clean_df, \"Condition\", value_vars3, replacements3)\n",
    "clean_tand = prepare_dataframe(clean_df, \"Condition\", value_vars4, replacements4)\n",
    "\n",
    "total_osc = prepare_dataframe(total_df, \"Condition\", value_vars1, replacements1)\n",
    "total_n = prepare_dataframe(total_df, \"Condition\", value_vars2, replacements2)                                  \n",
    "total_cop = prepare_dataframe(total_df, \"Condition\", value_vars3, replacements3)\n",
    "total_tand = prepare_dataframe(total_df, \"Condition\", value_vars4, replacements4)\n",
    "\n",
    "OA_total_osc = prepare_dataframe(OA_total_df, \"Blood\", value_vars1, replacements1)\n",
    "OA_total_osc = legend_prepare_dataframe(OA_total_osc, \"With\", \"Without\")\n",
    "OA_total_n = prepare_dataframe(OA_total_df, \"Blood\", value_vars2, replacements2)\n",
    "OA_total_n = legend_prepare_dataframe(OA_total_n, \"With\", \"Without\")\n",
    "OA_total_cop = prepare_dataframe(OA_total_df, \"Blood\", value_vars3, replacements3)\n",
    "OA_total_cop = legend_prepare_dataframe(OA_total_cop, \"With\", \"Without\")\n",
    "OA_total_tand = prepare_dataframe(OA_total_df, \"Blood\", value_vars4, replacements4)\n",
    "OA_total_tand = legend_prepare_dataframe(OA_total_tand, \"With\", \"Without\")\n",
    "\n",
    "OA_storage_osc = prepare_dataframe(OA_storage_df, \"Storage in freezer\", value_vars1, replacements1)\n",
    "OA_storage_osc = legend_prepare_dataframe(OA_storage_osc, \"Frozen\", \"Not Frozen\")\n",
    "OA_storage_n = prepare_dataframe(OA_storage_df, \"Storage in freezer\", value_vars2, replacements2)\n",
    "OA_storage_n = legend_prepare_dataframe(OA_storage_n, \"Frozen\", \"Not Frozen\")\n",
    "OA_storage_cop = prepare_dataframe(OA_storage_df, \"Storage in freezer\", value_vars3, replacements3)\n",
    "OA_storage_cop = legend_prepare_dataframe(OA_storage_cop, \"Frozen\", \"Not Frozen\")\n",
    "OA_storage_tand = prepare_dataframe(OA_storage_df, \"Storage in freezer\", value_vars4, replacements4)\n",
    "OA_storage_tand = legend_prepare_dataframe(OA_storage_tand, \"Frozen\", \"Not Frozen\")\n",
    "\n",
    "gender_OA_total_osc = prepare_dataframe(OA_total_df, \"Gender\", value_vars1, replacements1)\n",
    "gender_OA_total_n = prepare_dataframe(OA_total_df, \"Gender\", value_vars2, replacements2)\n",
    "gender_OA_total_cop = prepare_dataframe(OA_total_df, \"Gender\", value_vars3, replacements3)\n",
    "gender_OA_total_tand = prepare_dataframe(OA_total_df, \"Gender\", value_vars4, replacements4)\n",
    "\n",
    "# Perform t-tests and display results\n",
    "# Perform t-tests between healthy and OA dataframes\n",
    "column_names = [\n",
    "    \"Frequency Sweep G' at 0.68 rad/s (Pa)\",\n",
    "    \"Frequency Sweep G'' at 0.68 rad/s (Pa)\",\n",
    "    \"Time Sweep G' average (Pa)\",\n",
    "    \"Time Sweep G'' average (Pa)\",\n",
    "    \"Frequency Sweep G' at 14.58 rad/s (Pa)\",\n",
    "    \"Frequency Sweep G'' at 14.58 rad/s (Pa)\",\n",
    "    \"Flow Step Viscosity at 3.162 1/s (Pa.s)\",\n",
    "    \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\",\n",
    "    \"Cross Over Point (rad/s)\",\n",
    "    \"tan(delta) at 0.68 rad/s\",\n",
    "    \"tan(delta) at 3.142 rad/s\",\n",
    "    \"tan(delta) at 14.58 rad/s\"\n",
    "]\n",
    "\n",
    "# Perform t-tests and display results\n",
    "results_df = statistics(healthy_df_log, OA_df_log, column_names)\n",
    "display(results_df)\n",
    "\n",
    "# Define biomarker information\n",
    "biomarkers_info = [\n",
    "    # Biomarker name, corresponding column, limit, big_var\n",
    "    [\"G' at 0.68 rad/s\", \"Frequency Sweep G' at 0.68 rad/s (Pa)\", 2.18, \"control\"],\n",
    "    [\"G'' at 0.68 rad/s\", \"Frequency Sweep G'' at 0.68 rad/s (Pa)\", 1.88, \"control\"],\n",
    "    [\"G' at 3.142 rad/s\", \"Time Sweep G' average (Pa)\", 4.34, \"control\"],\n",
    "    [\"G'' at 3.142 rad/s\", \"Time Sweep G'' average (Pa)\", 2.58, \"control\"],\n",
    "    [\"G' at 14.58 rad/s\", \"Frequency Sweep G' at 14.58 rad/s (Pa)\", 7.28, \"control\"],\n",
    "    [\"G'' at 14.58 rad/s\", \"Frequency Sweep G'' at 14.58 rad/s (Pa)\", 3.06, \"control\"],\n",
    "    [\"Viscosity |n*| at 3.142 rad/s\", \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\", 1.65, \"control\"],\n",
    "    [\"tan(delta) at 0.68 rad/s\", \"tan(delta) at 0.68 rad/s\", 0.88, \"target\"],\n",
    "    [\"tan(delta) at 3.142 rad/s\", \"tan(delta) at 3.142 rad/s\", 0.65, \"target\"],\n",
    "    [\"tan(delta) at 14.58 rad/s\", \"tan(delta) at 14.58 rad/s\", 0.48, \"target\"]\n",
    "]\n",
    "biomarker_df = biomarker_validation(clean_df, biomarkers_info)\n",
    "display(biomarker_df)\n",
    "\n",
    "# Call functions to create custom boxplots\n",
    "# Create custom boxplots for oscillatory data\n",
    "create_custom_boxplot(clean_osc, \"gprgdpr_boxplot\", \"\", \"G' and G'' (Pa)\", [\"G'-0.68 rad/s\", \"G''-0.68 rad/s\", \"G'-3.142 rad/s\", \"G''-3.142 rad/s\", \"G'-14.58 rad/s\", \"G''-14.58 rad/s\"], \n",
    "                      hue=\"Condition\", hue_order=['healthy', 'oa'], legend_labels=['Healthy', 'OA'], \n",
    "                      width=0.8, dodge=True, yA=0, yB=None)\n",
    "\n",
    "# Create custom boxplots for tan(delta)\n",
    "create_custom_boxplot(clean_tand, \"tandelta_boxplot\", \"\", \"tan(δ)\", [\"0.68 rad/s\", \"3.142 rad/s\", \"14.58 rad/s\"],\n",
    "                      hue=\"Condition\", hue_order=['healthy', 'oa'], legend_labels=['Healthy', 'OA'], \n",
    "                      width=0.8, dodge=True, yA=0, yB=None)\n",
    "\n",
    "# Create custom boxplots for viscosity\n",
    "create_custom_boxplot(clean_n, \"viscosity_boxplot\", \"\", \"n and |n*| (Pa.s)\", [\"n\", \"|n*|\"],\n",
    "                      hue=\"Condition\", hue_order=['healthy', 'oa'], legend_labels=['Healthy', 'OA'], \n",
    "                      width=0.8, dodge=True, yA=0, yB=None)\n",
    "\n",
    "# Create custom boxplots for cross-over point\n",
    "create_custom_boxplot(clean_cop, \"crossover_boxplot\", \"\", \"Cross-over point (rad/s)\", [\"\"],\n",
    "                      hue=\"Condition\", hue_order=['healthy', 'oa'], legend_labels=['Healthy', 'OA'], \n",
    "                      width=0.2, dodge=True, yA=0, yB=None)\n",
    "\n",
    "# Perform multivariate Monte Carlo Simulation\n",
    "# Specify the columns to sample\n",
    "columns_to_sample = [\"Frequency Sweep G' at 0.68 rad/s (Pa)\",\n",
    "                     \"Frequency Sweep G'' at 0.68 rad/s (Pa)\",\n",
    "                     \"Time Sweep G' average (Pa)\",\n",
    "                     \"Time Sweep G'' average (Pa)\",\n",
    "                     \"Frequency Sweep G' at 14.58 rad/s (Pa)\",\n",
    "                     \"Frequency Sweep G'' at 14.58 rad/s (Pa)\",\n",
    "                     \"Frequency Sweep Viscosity at 3.142 rad/s (Pa.s)\"]\n",
    "\n",
    "# Sample from multivariate normal distribution for healthy_df_log\n",
    "healthy_artificial_df = multivariate_normal_sampling(healthy_df_log, columns_to_sample, scaled_factor=2)\n",
    "healthy_artificial_df['Condition'] = 0\n",
    "\n",
    "# Sample from multivariate normal distribution for OA_df_log\n",
    "OA_artificial_df = multivariate_normal_sampling(OA_df_log, columns_to_sample, scaled_factor=2)\n",
    "OA_artificial_df['Condition'] = 1\n",
    "\n",
    "# Concatenate the sampled DataFrames for both conditions\n",
    "artificial_df = pd.concat([healthy_artificial_df, OA_artificial_df], axis=0, ignore_index=True)\n",
    "\n",
    "# Shuffle the rows randomly\n",
    "artificial_df = artificial_df.reindex(np.random.permutation(artificial_df.index))\n",
    "\n",
    "# Reset the index of the DataFrame\n",
    "artificial_df = artificial_df.reset_index(drop=True)\n",
    "\n",
    "# Display the resulting artificial DataFrame\n",
    "display(artificial_df)\n",
    "\n",
    "# Train and evaluate a Logistics Regression model\n",
    "# Specify the columns to use as predictor variables\n",
    "columns_to_predict = [\"Time Sweep G' average (Pa)\"\n",
    "                      ]\n",
    "\n",
    "# Train, evaluate, and perform ROC analysis for the logistic regression model\n",
    "mean_cross_val_accuracy = logistic_regression(artificial_df, columns_to_predict)\n",
    "\n",
    "#print(\"Mean Cross-Validation Accuracy:\", mean_cross_val_accuracy)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3d239e7-729e-4fe2-a2e6-3c037e5636f8",
   "metadata": {},
   "source": [
    "Plot the Monte Carlo points."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d990a512",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 1.0)"
      ]
     },
     "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": [
    "artificial_df = artificial_df.copy().head(500)\n",
    "artificial_df[\"Condition\"].replace(0, \"Healthy\", inplace=True)\n",
    "artificial_df[\"Condition\"].replace(1, \"OA\", inplace=True)\n",
    "sns.scatterplot(data=artificial_df, x=\"Time Sweep G' average (Pa)\", y=\"Time Sweep G'' average (Pa)\", hue=\"Condition\")\n",
    "plt.xlim(0,1.2)\n",
    "plt.ylim(0,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f9a9527d-1186-4337-bb12-6759d79816c7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}