literature_review.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How to: Create a reproducible computational workflow using Jupyter notebook\n",
    "Content under Creative Commons Attribution license CC-BY-NC-SA 4.0   \n",
    "Code under GNU-GPL v3 License  \n",
    "© 2019 Serena Bonaretti\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook is an example of how to create a reproducible workflow using Jupyter notebook and python.  \n",
    "There are 4 steps that make this workflow reproducible:  \n",
    "1. [Automatically download data from a repository](#download)  \n",
    "2. [Automate data manipulation](#manipulation)\n",
    "3. [Define seeds to generate random numbers](#seeds)\n",
    "4. [Print dependences](#dependencies) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The example coded in this notebook consists in the creation of an interactive literature review map of femoral knee cartilage segmentation. The map shows type of algorithm, location, first author and year of publication.\n",
    "The input data is a `.csv` file on Zenodo (www.doi.org/10.5281/zenodo.3553483) containing data about the literature of femoral knee cartilage segmentation. \n",
    "There are 4 versions of this file on Zenodo that are used for three different YouTube videos: \n",
    "- Version 1 and 2: Used in the video [How to: Zenodo](https://youtu.be/S1qK_TA52e4)  \n",
    "- Version 2: Used in the video [How to: Reproducible computational workflows](https://youtu.be/-9qSUJTuec8)\n",
    "- Version 3: Not used in any video as it contains some empty columns added by mistake\n",
    "- Version 4: Used in the video [How to: binder]()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import wget # to download from zenodo\n",
    "import pandas as pd  \n",
    "import numpy  as np\n",
    "# from   geopy.geocoders import Nominatim\n",
    "import altair as alt\n",
    "import vega \n",
    "from   vega_datasets import data # for state contours "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "alt.renderers.enable('notebook'); # for rendering in jupyter notebook"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a name = \"download\"></a>\n",
    "## 1. Automatically download data from a repository\n",
    "\n",
    "Input data should be in a repository that provides a **persistent digital object identifier (DOI)** so that data will be available in the future. It is *discouraged* to share data from **personal repositories** because links tend to get deleted, thus compromising the reproducibility of the workflow.\n",
    "- Download the file `cart_segm_literature.csv` from Zenodo, open the downloaded file, and show the content\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# file name and zenodo url\n",
    "file_name  = \"cart_segm_literature.csv\"\n",
    "zenodo_url = \"https://zenodo.org/record/3553483/files/\" # 3553483 are the last digits of the specific version of dataset DOI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./cart_segm_literature.csv'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# download\n",
    "wget.download(zenodo_url + file_name, \"./\" + file_name) # input, output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Load the file `cart_segm_literature.csv` \n",
    "- Show data in file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load literature table\n",
    "literature = pd.read_csv(\"./\" + file_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>algorithm_type</th>\n",
       "      <th>bibtex_id</th>\n",
       "      <th>author_1</th>\n",
       "      <th>author_2</th>\n",
       "      <th>author_3</th>\n",
       "      <th>author_4</th>\n",
       "      <th>author_5</th>\n",
       "      <th>author6</th>\n",
       "      <th>author_7</th>\n",
       "      <th>author_8</th>\n",
       "      <th>affiliation_last_author</th>\n",
       "      <th>city_last_author</th>\n",
       "      <th>country_last_author</th>\n",
       "      <th>title</th>\n",
       "      <th>journal</th>\n",
       "      <th>year</th>\n",
       "      <th>link_to_pubmed</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>active contours</td>\n",
       "      <td>Solloway_1997</td>\n",
       "      <td>﻿Solloway S</td>\n",
       "      <td>Hutchinson CE</td>\n",
       "      <td>Waterton JC</td>\n",
       "      <td>Taylor CJ</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Department of Medical Biophysics, University o...</td>\n",
       "      <td>Manchester</td>\n",
       "      <td>UK</td>\n",
       "      <td>﻿The use of active shape models for making thi...</td>\n",
       "      <td>﻿Magn Reson Med 37(6):943–952</td>\n",
       "      <td>1997</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pubmed/9178247</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>-2.245115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>active contours</td>\n",
       "      <td>Vincent_2011</td>\n",
       "      <td>﻿Vincent G</td>\n",
       "      <td>Wolstenholme C</td>\n",
       "      <td>Scott I</td>\n",
       "      <td>Bowes M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Imorphics Ltd.</td>\n",
       "      <td>Manchester</td>\n",
       "      <td>UK</td>\n",
       "      <td>﻿Fully Automatic Segmentation of the Knee Join...</td>\n",
       "      <td>﻿MICCAI 2010 Workshop Medical Image Analysis f...</td>\n",
       "      <td>2011</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4...</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>-2.245115</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>...</td>\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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>hybrid</td>\n",
       "      <td>Seim_2010</td>\n",
       "      <td>Seim H</td>\n",
       "      <td>Kainmueller D</td>\n",
       "      <td>Lamecker H</td>\n",
       "      <td>Bindernagel M</td>\n",
       "      <td>Malinowski J</td>\n",
       "      <td>Zachow S</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Zuse Institute Berlin (ZIB), Medical Planning ...</td>\n",
       "      <td>Berlin</td>\n",
       "      <td>Germany</td>\n",
       "      <td>﻿Model-based auto-segmentation of knee bones a...</td>\n",
       "      <td>﻿Medical Image Analysis for the Clinic: A Gran...</td>\n",
       "      <td>2010</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52.517037</td>\n",
       "      <td>13.388860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>hybrid</td>\n",
       "      <td>Dam_2015</td>\n",
       "      <td>Dam ﻿E B</td>\n",
       "      <td>Lillholm M</td>\n",
       "      <td>Marques J</td>\n",
       "      <td>Nielsen M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Biomediq and University of Copenhagen</td>\n",
       "      <td>Copenhagen</td>\n",
       "      <td>Denmark</td>\n",
       "      <td>Automatic segmentation of high- and low-field ...</td>\n",
       "      <td>﻿SPIE  Journal of Medical Imaging, vol. 2, pp....</td>\n",
       "      <td>2015</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pubmed/26158096</td>\n",
       "      <td>55.686724</td>\n",
       "      <td>12.570072</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>29 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     algorithm_type      bibtex_id     author_1         author_2  \\\n",
       "0   active contours  Solloway_1997  ﻿Solloway S    Hutchinson CE   \n",
       "1   active contours   Vincent_2011   ﻿Vincent G   Wolstenholme C   \n",
       "..              ...            ...          ...              ...   \n",
       "27           hybrid      Seim_2010       Seim H    Kainmueller D   \n",
       "28           hybrid       Dam_2015     Dam ﻿E B       Lillholm M   \n",
       "\n",
       "        author_3        author_4       author_5    author6 author_7 author_8  \\\n",
       "0    Waterton JC       Taylor CJ            NaN        NaN      NaN      NaN   \n",
       "1        Scott I         Bowes M            NaN        NaN      NaN      NaN   \n",
       "..           ...             ...            ...        ...      ...      ...   \n",
       "27    Lamecker H   Bindernagel M   Malinowski J   Zachow S      NaN      NaN   \n",
       "28     Marques J       Nielsen M            NaN        NaN      NaN      NaN   \n",
       "\n",
       "                              affiliation_last_author city_last_author  \\\n",
       "0   Department of Medical Biophysics, University o...       Manchester   \n",
       "1                                      Imorphics Ltd.       Manchester   \n",
       "..                                                ...              ...   \n",
       "27  Zuse Institute Berlin (ZIB), Medical Planning ...           Berlin   \n",
       "28              Biomediq and University of Copenhagen       Copenhagen   \n",
       "\n",
       "   country_last_author                                              title  \\\n",
       "0                   UK  ﻿The use of active shape models for making thi...   \n",
       "1                   UK  ﻿Fully Automatic Segmentation of the Knee Join...   \n",
       "..                 ...                                                ...   \n",
       "27             Germany  ﻿Model-based auto-segmentation of knee bones a...   \n",
       "28             Denmark  Automatic segmentation of high- and low-field ...   \n",
       "\n",
       "                                              journal  year  \\\n",
       "0                       ﻿Magn Reson Med 37(6):943–952  1997   \n",
       "1   ﻿MICCAI 2010 Workshop Medical Image Analysis f...  2011   \n",
       "..                                                ...   ...   \n",
       "27  ﻿Medical Image Analysis for the Clinic: A Gran...  2010   \n",
       "28  ﻿SPIE  Journal of Medical Imaging, vol. 2, pp....  2015   \n",
       "\n",
       "                                       link_to_pubmed   latitude  longitude  \n",
       "0         https://www.ncbi.nlm.nih.gov/pubmed/9178247  53.479489  -2.245115  \n",
       "1   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4...  53.479489  -2.245115  \n",
       "..                                                ...        ...        ...  \n",
       "27                                                NaN  52.517037  13.388860  \n",
       "28       https://www.ncbi.nlm.nih.gov/pubmed/26158096  55.686724  12.570072  \n",
       "\n",
       "[29 rows x 19 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# show data\n",
    "\n",
    "# display all rows and columns\n",
    "dataDimension = literature.shape # get number of rows\n",
    "pd.set_option(\"display.max_rows\",5)\n",
    "pd.set_option(\"display.max_columns\",dataDimension[1])\n",
    "\n",
    "# show\n",
    "literature"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a name = \"manipulation\"></a>\n",
    "## 2. Automate data manipulation\n",
    "\n",
    "**Automatic** data manipulation does not compromise original data and keeps track of manipulations, making analyses reproducible. It is *discouraged* to do **manual** manipulation, as it compromizes original data, is prone to errors, and does not keep track of changes, making analyses hardly reproducible.\n",
    "\n",
    "- Change `bibtext_id` format from `author_year` to `author (year)` for better readability when hovering (e.g. from `Solloway_1997` to `Solloway (1997)`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>algorithm_type</th>\n",
       "      <th>bibtex_id</th>\n",
       "      <th>author_1</th>\n",
       "      <th>author_2</th>\n",
       "      <th>author_3</th>\n",
       "      <th>author_4</th>\n",
       "      <th>author_5</th>\n",
       "      <th>author6</th>\n",
       "      <th>author_7</th>\n",
       "      <th>author_8</th>\n",
       "      <th>affiliation_last_author</th>\n",
       "      <th>city_last_author</th>\n",
       "      <th>country_last_author</th>\n",
       "      <th>title</th>\n",
       "      <th>journal</th>\n",
       "      <th>year</th>\n",
       "      <th>link_to_pubmed</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>active contours</td>\n",
       "      <td>Solloway (1997)</td>\n",
       "      <td>﻿Solloway S</td>\n",
       "      <td>Hutchinson CE</td>\n",
       "      <td>Waterton JC</td>\n",
       "      <td>Taylor CJ</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Department of Medical Biophysics, University o...</td>\n",
       "      <td>Manchester</td>\n",
       "      <td>UK</td>\n",
       "      <td>﻿The use of active shape models for making thi...</td>\n",
       "      <td>﻿Magn Reson Med 37(6):943–952</td>\n",
       "      <td>1997</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pubmed/9178247</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>-2.245115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>active contours</td>\n",
       "      <td>Vincent (2011)</td>\n",
       "      <td>﻿Vincent G</td>\n",
       "      <td>Wolstenholme C</td>\n",
       "      <td>Scott I</td>\n",
       "      <td>Bowes M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Imorphics Ltd.</td>\n",
       "      <td>Manchester</td>\n",
       "      <td>UK</td>\n",
       "      <td>﻿Fully Automatic Segmentation of the Knee Join...</td>\n",
       "      <td>﻿MICCAI 2010 Workshop Medical Image Analysis f...</td>\n",
       "      <td>2011</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4...</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>-2.245115</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>...</td>\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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>hybrid</td>\n",
       "      <td>Seim (2010)</td>\n",
       "      <td>Seim H</td>\n",
       "      <td>Kainmueller D</td>\n",
       "      <td>Lamecker H</td>\n",
       "      <td>Bindernagel M</td>\n",
       "      <td>Malinowski J</td>\n",
       "      <td>Zachow S</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Zuse Institute Berlin (ZIB), Medical Planning ...</td>\n",
       "      <td>Berlin</td>\n",
       "      <td>Germany</td>\n",
       "      <td>﻿Model-based auto-segmentation of knee bones a...</td>\n",
       "      <td>﻿Medical Image Analysis for the Clinic: A Gran...</td>\n",
       "      <td>2010</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52.517037</td>\n",
       "      <td>13.388860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>hybrid</td>\n",
       "      <td>Dam (2015)</td>\n",
       "      <td>Dam ﻿E B</td>\n",
       "      <td>Lillholm M</td>\n",
       "      <td>Marques J</td>\n",
       "      <td>Nielsen M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Biomediq and University of Copenhagen</td>\n",
       "      <td>Copenhagen</td>\n",
       "      <td>Denmark</td>\n",
       "      <td>Automatic segmentation of high- and low-field ...</td>\n",
       "      <td>﻿SPIE  Journal of Medical Imaging, vol. 2, pp....</td>\n",
       "      <td>2015</td>\n",
       "      <td>https://www.ncbi.nlm.nih.gov/pubmed/26158096</td>\n",
       "      <td>55.686724</td>\n",
       "      <td>12.570072</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>29 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     algorithm_type        bibtex_id     author_1         author_2  \\\n",
       "0   active contours  Solloway (1997)  ﻿Solloway S    Hutchinson CE   \n",
       "1   active contours   Vincent (2011)   ﻿Vincent G   Wolstenholme C   \n",
       "..              ...              ...          ...              ...   \n",
       "27           hybrid      Seim (2010)       Seim H    Kainmueller D   \n",
       "28           hybrid       Dam (2015)     Dam ﻿E B       Lillholm M   \n",
       "\n",
       "        author_3        author_4       author_5    author6 author_7 author_8  \\\n",
       "0    Waterton JC       Taylor CJ            NaN        NaN      NaN      NaN   \n",
       "1        Scott I         Bowes M            NaN        NaN      NaN      NaN   \n",
       "..           ...             ...            ...        ...      ...      ...   \n",
       "27    Lamecker H   Bindernagel M   Malinowski J   Zachow S      NaN      NaN   \n",
       "28     Marques J       Nielsen M            NaN        NaN      NaN      NaN   \n",
       "\n",
       "                              affiliation_last_author city_last_author  \\\n",
       "0   Department of Medical Biophysics, University o...       Manchester   \n",
       "1                                      Imorphics Ltd.       Manchester   \n",
       "..                                                ...              ...   \n",
       "27  Zuse Institute Berlin (ZIB), Medical Planning ...           Berlin   \n",
       "28              Biomediq and University of Copenhagen       Copenhagen   \n",
       "\n",
       "   country_last_author                                              title  \\\n",
       "0                   UK  ﻿The use of active shape models for making thi...   \n",
       "1                   UK  ﻿Fully Automatic Segmentation of the Knee Join...   \n",
       "..                 ...                                                ...   \n",
       "27             Germany  ﻿Model-based auto-segmentation of knee bones a...   \n",
       "28             Denmark  Automatic segmentation of high- and low-field ...   \n",
       "\n",
       "                                              journal  year  \\\n",
       "0                       ﻿Magn Reson Med 37(6):943–952  1997   \n",
       "1   ﻿MICCAI 2010 Workshop Medical Image Analysis f...  2011   \n",
       "..                                                ...   ...   \n",
       "27  ﻿Medical Image Analysis for the Clinic: A Gran...  2010   \n",
       "28  ﻿SPIE  Journal of Medical Imaging, vol. 2, pp....  2015   \n",
       "\n",
       "                                       link_to_pubmed   latitude  longitude  \n",
       "0         https://www.ncbi.nlm.nih.gov/pubmed/9178247  53.479489  -2.245115  \n",
       "1   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4...  53.479489  -2.245115  \n",
       "..                                                ...        ...        ...  \n",
       "27                                                NaN  52.517037  13.388860  \n",
       "28       https://www.ncbi.nlm.nih.gov/pubmed/26158096  55.686724  12.570072  \n",
       "\n",
       "[29 rows x 19 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# replace underscore with space and opening bracket\n",
    "literature[\"bibtex_id\"] = literature[\"bibtex_id\"].str.replace('_',' (')\n",
    "# adding closing bracket\n",
    "literature[\"bibtex_id\"] = literature[\"bibtex_id\"].astype(str) + \")\"  \n",
    "\n",
    "# show table\n",
    "literature"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<a name = \"seeds\"></a>\n",
    "## 3. Define seeds to generate random numbers\n",
    "\n",
    "Algorithms create random numbers starting from a seed number (usually picked from the system clock) and following a set of instructions that output the wanted random number. **Fixing the seed number** allows to always compute the same random number.\n",
    "\n",
    "- Calculate latitude and longitude of `city_last_author` and add small random number to avoid overlap on map\n",
    "- *Note*: The cell below contains the commands to compute `latitude` and `longitude` from `city_last_author` using the `geopy` package. The cell is commented out because `geopy` often times out, affecting the execution of the notebook. For this reason, the columns `latitude` and `longitude` are already included in version 4 of the file `cart_segm_literature.csv` on Zenodo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate latitute and longitude \n",
    "# (timeout error can happen. wait a bit, and rerun the cell)\n",
    "# nom = Nominatim(user_agent=\"a_name\") # specify a name to avoid timeout errors\n",
    "\n",
    "# add latitude and longitude to literature\n",
    "# literature[\"coordinates\"] = literature[\"city_last_author\"].apply(nom.geocode, timeout=5) # create the intermediate columm \"coordinates\" to take advantage of pandas apply() method\n",
    "# literature[\"latitude\"]    = literature[\"coordinates\"].apply(lambda x: x.lat if x!=None else None) # take advantage of pandas apply() method to avoid for loop\n",
    "# literature[\"longitude\"]   = literature[\"coordinates\"].apply(lambda x: x.lon if x!=None else None) \n",
    "# literature = literature.drop(columns=\"coordinates\") # delete the column \"coordinates\" (altair will not be able to read it (even if not used))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>latitude</th>\n",
       "      <th>latitude_random</th>\n",
       "      <th>longitude</th>\n",
       "      <th>longitude_random</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>54.856484</td>\n",
       "      <td>-2.245115</td>\n",
       "      <td>0.095844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>55.686724</td>\n",
       "      <td>56.656652</td>\n",
       "      <td>12.570072</td>\n",
       "      <td>14.436938</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>29 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     latitude  latitude_random  longitude  longitude_random\n",
       "0   53.479489        54.856484  -2.245115          0.095844\n",
       "..        ...              ...        ...               ...\n",
       "28  55.686724        56.656652  12.570072         14.436938\n",
       "\n",
       "[29 rows x 4 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# FIRST RUN\n",
    "np.random.seed(seed=3) # if this is not present, the cell is not reproducible\n",
    "\n",
    "# add randomness to latitude\n",
    "random_lat = np.random.uniform(low=0.0, high=2.5, size=(len(literature[\"latitude\"]),))\n",
    "literature[\"latitude_random\"] = literature[\"latitude\"] + pd.Series(random_lat)\n",
    "\n",
    "# add randomness to longitude\n",
    "random_lon = np.random.uniform(low=0.0, high=2.5, size=(len(literature[\"longitude\"]),))\n",
    "literature[\"longitude_random\"] = literature[\"longitude\"] + pd.Series(random_lon)\n",
    "\n",
    "# show \n",
    "pd.set_option(\"display.max_rows\",3) # show only the first three rows\n",
    "literature [[\"latitude\", \"latitude_random\", \"longitude\", \"longitude_random\"]] # show only these columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>latitude</th>\n",
       "      <th>latitude_random</th>\n",
       "      <th>longitude</th>\n",
       "      <th>longitude_random</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>53.479489</td>\n",
       "      <td>54.856484</td>\n",
       "      <td>-2.245115</td>\n",
       "      <td>0.095844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>55.686724</td>\n",
       "      <td>56.656652</td>\n",
       "      <td>12.570072</td>\n",
       "      <td>14.436938</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>29 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     latitude  latitude_random  longitude  longitude_random\n",
       "0   53.479489        54.856484  -2.245115          0.095844\n",
       "..        ...              ...        ...               ...\n",
       "28  55.686724        56.656652  12.570072         14.436938\n",
       "\n",
       "[29 rows x 4 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SECOND RUN\n",
    "np.random.seed(seed=3) # if this is not present, the cell is not reproducible\n",
    "\n",
    "# add randomness to latitude\n",
    "random_lat = np.random.uniform(low=0.0, high=2.5, size=(len(literature[\"latitude\"]),))\n",
    "literature[\"latitude_random\"] = literature[\"latitude\"] + pd.Series(random_lat)\n",
    "\n",
    "# add randomness to longitude\n",
    "random_lon = np.random.uniform(low=0.0, high=2.5, size=(len(literature[\"longitude\"]),))\n",
    "literature[\"longitude_random\"] = literature[\"longitude\"] + pd.Series(random_lon)\n",
    "\n",
    "# show literature\n",
    "pd.set_option(\"display.max_rows\",3) # show only the first three rows\n",
    "literature [[\"latitude\", \"latitude_random\", \"longitude\", \"longitude_random\"]] # show only these columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize map\n",
    "- Show the interactive map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v3+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v3.4.0.json",
       "config": {
        "mark": {
         "tooltip": null
        },
        "view": {
         "height": 300,
         "width": 400
        }
       },
       "datasets": {
        "data-861c1f453e34c20f068f63b2a7d2c18f": [
         {
          "affiliation_last_author": "Department of Medical Biophysics, University of Manchester\n",
          "algorithm_type": "active contours",
          "author6": null,
          "author_1": "﻿Solloway S",
          "author_2": " Hutchinson CE",
          "author_3": " Waterton JC",
          "author_4": " Taylor CJ",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Solloway (1997)",
          "city_last_author": "Manchester",
          "country_last_author": "UK",
          "journal": "﻿Magn Reson Med 37(6):943–952",
          "latitude": 53.479489,
          "latitude_random": 54.85648375643644,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/9178247",
          "longitude": -2.245115,
          "longitude_random": 0.09584412465107572,
          "title": "﻿The use of active shape models for making thickness measure- ments of articular cartilage from MR images",
          "year": 1997
         },
         {
          "affiliation_last_author": "Imorphics Ltd.",
          "algorithm_type": "active contours",
          "author6": null,
          "author_1": "﻿Vincent G",
          "author_2": " Wolstenholme C",
          "author_3": " Scott I",
          "author_4": " Bowes M",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Vincent (2011)",
          "city_last_author": "Manchester",
          "country_last_author": "UK",
          "journal": "﻿MICCAI 2010 Workshop Medical Image Analysis for the Clinic—A Grand Challenge (SKI0)",
          "latitude": 53.479489,
          "latitude_random": 55.24985855654526,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4478858/",
          "longitude": -2.245115,
          "longitude_random": 0.19487355618233426,
          "title": "﻿Fully Automatic Segmentation of the Knee Joint using Active Appearance Models",
          "year": 2011
         },
         {
          "affiliation_last_author": "Imaging Science and Biomedical Engineering, School of\nCancer and Enabling Sciences, The University of Manchester",
          "algorithm_type": "active contours",
          "author6": " Moots  RJ",
          "author_1": "﻿Williams TG",
          "author_2": " Holmes AP",
          "author_3": " Waterton JC",
          "author_4": " Maciewicz RA",
          "author_5": " Hutchinson CE",
          "author_7": "Nash AFP",
          "author_8": " Taylor CJ",
          "bibtex_id": "Williams (2010)",
          "city_last_author": "Manchester",
          "country_last_author": "UK",
          "journal": "﻿IEEE transactions on medical imaging, vol. 29, pp. 1541-1559",
          "latitude": 53.479489,
          "latitude_random": 54.206750847282365,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/20378463",
          "longitude": -2.245115,
          "longitude_random": -0.564155810217966,
          "title": "﻿Anatomically corresponded regional analysis of cartilage in asymptomatic and osteoarthritic knees by statistical shape modelling of the bone",
          "year": 2010
         },
         {
          "affiliation_last_author": "Computer Science Department, University of Basel",
          "algorithm_type": "active contours",
          "author6": null,
          "author_1": "Amberg",
          "author_2": "Luthi M",
          "author_3": "Vetter T",
          "author_4": null,
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Amberg (2010)",
          "city_last_author": "Basel",
          "country_last_author": "Switzerland",
          "journal": null,
          "latitude": 47.558108000000004,
          "latitude_random": 48.83517701299416,
          "link_to_pubmed": null,
          "longitude": 7.587826,
          "longitude_random": 9.844911271345994,
          "title": "Fully Automated Segmentation of the Knee using Local Deformation-Model Fitting",
          "year": 2010
         },
         {
          "affiliation_last_author": "Department of Radiology and\nBiomedical Imaging, UC San Francisco",
          "algorithm_type": "active contours",
          "author6": " Link T",
          "author_1": "Carballido-Gamio J",
          "author_2": " Bauer J",
          "author_3": " Stahl R",
          "author_4": " Lee KY",
          "author_5": " Krause S",
          "author_7": " Majumdar S",
          "author_8": null,
          "bibtex_id": "CarballidoGamio (2008)",
          "city_last_author": "San Francisco",
          "country_last_author": "USA",
          "journal": "﻿Medical Image Analysis",
          "latitude": 37.779026,
          "latitude_random": 40.01139338586914,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/17923429",
          "longitude": -122.41990600000001,
          "longitude_random": -120.30552882176707,
          "title": "Inter-subject comparison of MRI knee cartilage thickness",
          "year": 2008
         },
         {
          "affiliation_last_author": "Qmetrics Technology, LLC",
          "algorithm_type": "atlas-based",
          "author6": " Schreyer E",
          "author_1": "﻿Tamez-Pena J",
          "author_2": " Gonzalez J",
          "author_3": " Farber J",
          "author_4": "Gonzales PC",
          "author_5": " Schreyer E",
          "author_7": " Totterman S",
          "author_8": null,
          "bibtex_id": "TamezPena (2012)",
          "city_last_author": "Rochester",
          "country_last_author": "USA",
          "journal": "IEEE Transactions on biomedical engineering",
          "latitude": 43.157284999999995,
          "latitude_random": 45.39801772233359,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/22318477",
          "longitude": -77.615214,
          "longitude_random": -76.67022889667777,
          "title": "Unsupervised Segmentation and Quantification\nof Anatomical Knee Features: Data From the\nOsteoarthritis Initiative",
          "year": 2012
         },
         {
          "affiliation_last_author": "Department of Computer Science, University of North Carolina",
          "algorithm_type": "atlas-based",
          "author6": null,
          "author_1": "﻿Shan L",
          "author_2": " Zach C",
          "author_3": " Charles C",
          "author_4": " Niethammer M",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Shan (2014)",
          "city_last_author": "Chapel Hill",
          "country_last_author": "USA",
          "journal": "﻿Med Image Anal 18(7):1233–1246",
          "latitude": 35.913154,
          "latitude_random": 36.22711727615959,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/25128683",
          "longitude": -79.05578,
          "longitude_random": -78.82523747782069,
          "title": "﻿Automatic atlas-based three-label cartilage segmentation from MR knee images",
          "year": 2014
         },
         {
          "affiliation_last_author": "Department of Radiology and\nBiomedical Imaging, UC San Francisco",
          "algorithm_type": "atlas-based",
          "author6": null,
          "author_1": "﻿Pedoia V",
          "author_2": " Li X",
          "author_3": " Su F",
          "author_4": " Calixto N",
          "author_5": " Majumdar S",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Pedoia (2015)",
          "city_last_author": "San Francisco",
          "country_last_author": "USA",
          "journal": "J Magn Reson Imaging Apr;43(4):970-80.",
          "latitude": 37.779026,
          "latitude_random": 38.29713319534547,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/26443990",
          "longitude": -122.41990600000001,
          "longitude_random": -120.78637874369117,
          "title": "﻿Fully automatic analysis of the knee articular cartilage T1ρ relaxation time using voxel-based relaxometry",
          "year": 2015
         },
         {
          "affiliation_last_author": "Yıldız Technical University, Computer Engineering Department",
          "algorithm_type": "graph-based",
          "author6": null,
          "author_1": "Öztürk ﻿CN",
          "author_2": " Albayrak S",
          "author_3": null,
          "author_4": null,
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Ozturk (2016)",
          "city_last_author": "Istanbul",
          "country_last_author": "Turkey",
          "journal": "﻿Computers in biology and medicine, vol. 72, pp. 90-107",
          "latitude": 41.009633,
          "latitude_random": 41.13830100825208,
          "link_to_pubmed": null,
          "longitude": 28.965165000000002,
          "longitude_random": 30.359766905971128,
          "title": "﻿Automatic segmentation of cartilage in high-field magnetic resonance images of the knee joint with an improved voxel-classification-driven region-growing algorithm using vicinity-correlated subsampling",
          "year": 2016
         },
         {
          "affiliation_last_author": "Department of Medical Imaging, College of Biomedical Engineering, Third Military Medical University",
          "algorithm_type": "graph-based",
          "author6": " Qiu M",
          "author_1": "Wang P",
          "author_2": "  He X",
          "author_3": " Li Y",
          "author_4": " Zhu X",
          "author_5": " Chen W",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "WangP (2016)",
          "city_last_author": "Chongqing",
          "country_last_author": "China",
          "journal": "﻿Journal of Medical Imaging and Health Informatics, vol. 6, pp. 948-956",
          "latitude": 29.558571,
          "latitude_random": 30.66059560912659,
          "link_to_pubmed": null,
          "longitude": 106.549282,
          "longitude_random": 107.45319390765646,
          "title": "﻿Automatic Knee Cartilage Segmentation Using Multi-Feature Support Vector Machine and Elastic Region Growing for Magnetic Resonance Images",
          "year": 2016
         },
         {
          "affiliation_last_author": "Department of Electrical and Computer Engineering, The University of Iowa",
          "algorithm_type": "graph-based",
          "author6": " Sonka M ",
          "author_1": "Yin Y",
          "author_2": "Zhang  X",
          "author_3": "Williams  R",
          "author_4": "Wu  X",
          "author_5": "Anderson  DD",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Yin (2010)",
          "city_last_author": "Iowa City",
          "country_last_author": "USA",
          "journal": "﻿IEEE transactions on medical imaging, vol. 29, pp. 2023-2037",
          "latitude": 41.661256,
          "latitude_random": 41.73594652719642,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/20643602",
          "longitude": -91.529911,
          "longitude_random": -90.96727473790042,
          "title": "﻿\"LOGISMOS—layered optimal graph image segmentation of multiple objects and surfaces: cartilage segmentation in the knee joint",
          "year": 2010
         },
         {
          "affiliation_last_author": "Section of Rheumatology, Pittsburgh VA Healthcare System",
          "algorithm_type": "graph-based",
          "author6": "Boudreau  R",
          "author_1": "Bae ﻿K",
          "author_2": "Shim  H",
          "author_3": "Tao  C",
          "author_4": "Chang  S",
          "author_5": "Wang  J",
          "author_7": "Kwoh CK",
          "author_8": null,
          "bibtex_id": "Bae (2009)",
          "city_last_author": "Pittsburgh",
          "country_last_author": "USA",
          "journal": "﻿Osteoarthritis and Cartilage, vol. 17, pp. 1589-1597",
          "latitude": 40.441694,
          "latitude_random": 41.583777060986776,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/19577672",
          "longitude": -79.990086,
          "longitude_random": -78.97378620905938,
          "title": "﻿Intra-and inter-observer reproducibility of volume measurement of kneecartilage segmented from the OAI MR image set using a novel semi-automated segmentation method",
          "year": 2009
         },
         {
          "affiliation_last_author": "Department of Radiology, University of Pittsburgh School of\nMedicine",
          "algorithm_type": "graph-based",
          "author6": " Bae KT",
          "author_1": "﻿Shim H",
          "author_2": " Chang S",
          "author_3": " Tao C",
          "author_4": " Wang JH",
          "author_5": " Kwoh CK",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Shim (2009)",
          "city_last_author": "Pittsburgh",
          "country_last_author": "USA",
          "journal": "﻿Radiology 251(2):548–556",
          "latitude": 40.441694,
          "latitude_random": 42.0645541190369,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/19401579",
          "longitude": -79.990086,
          "longitude_random": -78.81773537825208,
          "title": "﻿Knee cartilage: Efficient and reproducible segmentation on high-spatial-resolution MR images with the semiautomated graph-cut algorithm method",
          "year": 2009
         },
         {
          "affiliation_last_author": "Center for Clinical and Basic\nResearch",
          "algorithm_type": "machine learning",
          "author6": null,
          "author_1": "﻿Folkesson J",
          "author_2": " Dam EB",
          "author_3": " Olsen OF",
          "author_4": " Pettersen PC",
          "author_5": " Christiansen C",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Folkesson (2007)",
          "city_last_author": "Ballerup",
          "country_last_author": "Denmark",
          "journal": "﻿IEEE Trans Med Imaging 26(1):106–115",
          "latitude": 55.730359,
          "latitude_random": 56.42657720661994,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/17243589",
          "longitude": 12.361231,
          "longitude_random": 13.034319945627653,
          "title": "﻿Segmenting articular cartilage automatically using a voxel classification approach",
          "year": 2007
         },
         {
          "affiliation_last_author": "Biomediq",
          "algorithm_type": "machine learning",
          "author6": " Nielsen M",
          "author_1": "﻿Prasoon A",
          "author_2": " Igel C",
          "author_3": " Loog M",
          "author_4": " Lauze F",
          "author_5": " Dam EB",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Prasoon (2013)",
          "city_last_author": "Copenhagen",
          "country_last_author": "Denmark",
          "journal": "﻿(EMBC), 35th Annual International Conference of the IEEE, pp 5469–5472",
          "latitude": 55.686724,
          "latitude_random": 57.37736125495033,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/24110974",
          "longitude": 12.570072,
          "longitude_random": 13.299553935563582,
          "title": "Femoral cartilage segmentation in Knee MRI scans using two stage voxel classification",
          "year": 2013
         },
         {
          "affiliation_last_author": "Department of Radiology and BRIC, University of North Carolina at Chapel Hill",
          "algorithm_type": "machine learning",
          "author6": null,
          "author_1": "Liu Q",
          "author_2": " Wang Q",
          "author_3": " Zhang L",
          "author_4": " Gao Y",
          "author_5": " Shen D",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Liu (2015)",
          "city_last_author": "Chapel Hill",
          "country_last_author": "USA",
          "journal": "﻿International Workshop on Machine Learning in Medical Imaging, pp. 186-193",
          "latitude": 35.913154,
          "latitude_random": 37.390311043540876,
          "link_to_pubmed": null,
          "longitude": -79.05578,
          "longitude_random": -77.91156400169255,
          "title": "﻿Multi-atlas context forests for knee MR image segmentation",
          "year": 2015
         },
         {
          "affiliation_last_author": "School of Electrical and Electronic Engineering, Nanyang Technological University",
          "algorithm_type": "machine learning",
          "author6": null,
          "author_1": "﻿Zhang K",
          "author_2": " Lu W",
          "author_3": " Marziliano P",
          "author_4": null,
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Zhang (2013)",
          "city_last_author": "Singapore",
          "country_last_author": "Singapore",
          "journal": "﻿Magn Reson Imaging 31(10):1731–1743",
          "latitude": 1.340863,
          "latitude_random": 1.4008177059429134,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/23867282",
          "longitude": 103.830392,
          "longitude_random": 105.98172678236732,
          "title": "﻿Automatic knee cartilage segmentation from multi-contrast MR images using support vector machine classification with spatial dependencies",
          "year": 2013
         },
         {
          "affiliation_last_author": "Department of Computer Science, College of Biomedical Engineering, Third Military Medical University",
          "algorithm_type": "machine learning",
          "author6": null,
          "author_1": "﻿Pang J",
          "author_2": " Li P",
          "author_3": " Qiu M",
          "author_4": " Chen W",
          "author_5": " Qiao L",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Pang (2015)",
          "city_last_author": "Chongqing",
          "country_last_author": "China",
          "journal": "J Digit Imaging. Dec; 28(6): 695–703",
          "latitude": 29.558571,
          "latitude_random": 30.955706219977206,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/25700618",
          "longitude": 106.549282,
          "longitude_random": 108.01491426116957,
          "title": "﻿Automatic articular cartilage segmentation based on pattern recognition from knee MRI images",
          "year": 2015
         },
         {
          "affiliation_last_author": "Biomediq",
          "algorithm_type": "deep learning",
          "author6": " Nielsen M",
          "author_1": "﻿Prasoon A",
          "author_2": " Petersen K",
          "author_3": " Igel C",
          "author_4": " Lauze F",
          "author_5": " Dam E",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Prasoon (2013a)",
          "city_last_author": "Copenhagen",
          "country_last_author": "Denmark",
          "journal": "﻿Med Image Comput Comput Assist Interv 16:246–253",
          "latitude": 55.686724,
          "latitude_random": 56.33485511726866,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/24579147",
          "longitude": 12.570072,
          "longitude_random": 13.278791656080525,
          "title": "﻿Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network",
          "year": 2013
         },
         {
          "affiliation_last_author": "Department of Radiology and Biomedical Imaging and Center for Digital Health Innovation (CDHI) University of California",
          "algorithm_type": "deep learning",
          "author6": null,
          "author_1": "Norman B",
          "author_2": " Pedoia V.",
          "author_3": " Majumdar S.",
          "author_4": null,
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Norman (2018)",
          "city_last_author": "San Francisco",
          "country_last_author": "USA",
          "journal": "Radiology. Jul;288(1):177-185",
          "latitude": 37.779026,
          "latitude_random": 38.81677899252517,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/29584598",
          "longitude": -122.41990600000001,
          "longitude_random": -121.724962233441,
          "title": "Use of 2D U-Net Convolutional Neural Networks for Automated Cartilage and Meniscus Segmentation of Knee MR Imaging Data to Determine Relaxometry and Morphometry",
          "year": 2018
         },
         {
          "affiliation_last_author": "Department of Radiology, University of Wisconsin School of Medicine and Public Health",
          "algorithm_type": "deep learning",
          "author6": null,
          "author_1": "Zhou Z",
          "author_2": " Zhao G",
          "author_3": " Kijowski R",
          "author_4": " Liu F",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Zhou (2018)",
          "city_last_author": "Madison",
          "country_last_author": "USA",
          "journal": "Magn Reson Med. 2018",
          "latitude": 43.074760999999995,
          "latitude_random": 43.78357370442829,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/28733975",
          "longitude": -89.38376099999999,
          "longitude_random": -88.24720581119513,
          "title": "Deep convolutional neural network for segmentation of knee joint anatomy",
          "year": 2018
         },
         {
          "affiliation_last_author": "Department of Radiology, University of Wisconsin School of Medicine and Public Health",
          "algorithm_type": "deep learning",
          "author6": " Kijowski R1",
          "author_1": "Liu F",
          "author_2": " Zhou Z",
          "author_3": " Jang H",
          "author_4": " Samsonov A",
          "author_5": " Zhao G1",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Liu (2018)",
          "city_last_author": "Madison",
          "country_last_author": "USA",
          "journal": "Magn Reson Med. Apr;79(4):2379-2391.",
          "latitude": 43.074760999999995,
          "latitude_random": 44.807605795782486,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/28733975",
          "longitude": -89.38376099999999,
          "longitude_random": -88.87023513848753,
          "title": "Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging.",
          "year": 2018
         },
         {
          "affiliation_last_author": "Department of Electrical Engineering, Seoul National University",
          "algorithm_type": "hybrid",
          "author6": null,
          "author_1": "﻿Lee S",
          "author_2": " Park SH",
          "author_3": " Shim H",
          "author_4": " Yun D",
          "author_5": " Lee S",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "LeeS (2011)",
          "city_last_author": "Seoul",
          "country_last_author": "Republic of Korea",
          "journal": "﻿Comput Vis Image Underst 115(12):1710–1720",
          "latitude": 37.566679,
          "latitude_random": 38.66781329417685,
          "link_to_pubmed": null,
          "longitude": 126.978291,
          "longitude_random": 127.48173777760768,
          "title": "﻿Optimization of local shape and appearance probabilities for segmentation of knee cartilage in 3-D MR images",
          "year": 2011
         },
         {
          "affiliation_last_author": "Department of Computing, Imperial College London ",
          "algorithm_type": "hybrid",
          "author6": null,
          "author_1": "﻿Wang Z",
          "author_2": " Donoghue C",
          "author_3": " Rueckert D",
          "author_4": null,
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "WangZ (2013)",
          "city_last_author": "London",
          "country_last_author": "UK",
          "journal": "lecture notes in computer science machine learning in medical. Imaging 8184(2013):98–105",
          "latitude": 51.507321999999995,
          "latitude_random": 51.8994913461874,
          "link_to_pubmed": null,
          "longitude": -0.127647,
          "longitude_random": 1.1574406495436047,
          "title": "Patch-based segmentation without registration: Application to knee MRI",
          "year": 2013
         },
         {
          "affiliation_last_author": "Department of Radiology, University of Pittsburgh",
          "algorithm_type": "hybrid",
          "author6": null,
          "author_1": "﻿Lee JG",
          "author_2": " Gumus S",
          "author_3": " Moon CH",
          "author_4": " Kwoh CK",
          "author_5": " Bae KT",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "LeeJG (2014)",
          "city_last_author": "Pittsburgh",
          "country_last_author": "USA",
          "journal": "﻿Med Phys 41(9):092303",
          "latitude": 40.441694,
          "latitude_random": 41.80331654507961,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/25186408",
          "longitude": -79.990086,
          "longitude_random": -79.77201257804799,
          "title": "﻿Fully automated segmentation of cartilage from the MR images of knee using a multi-atlas and local structural analysis method",
          "year": 2014
         },
         {
          "affiliation_last_author": "Siemens Corporate Research",
          "algorithm_type": "hybrid",
          "author6": " Zhou SK",
          "author_1": "﻿Wang Q",
          "author_2": " Wu D",
          "author_3": " Lu L",
          "author_4": " Liu M",
          "author_5": " Boyer KL",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "WangQ (2013)",
          "city_last_author": "Princeton",
          "country_last_author": "USA",
          "journal": "﻿Medical Computer Vision. Large Data in Medical Imaging Lecture Notes in Computer Sci- ence, vol 8331. Springer, Newyork, pp 105–115",
          "latitude": 38.355323999999996,
          "latitude_random": 40.30611091127842,
          "link_to_pubmed": null,
          "longitude": -87.567522,
          "longitude_random": -86.35855816912103,
          "title": "﻿Semantic Context Forests for Learning-Based Knee Cartilage Segmentation in 3D MR Images",
          "year": 2014
         },
         {
          "affiliation_last_author": "Zuse Institute Berlin and  1000shapes GmbH",
          "algorithm_type": "hybrid",
          "author6": null,
          "author_1": "Ambellan F",
          "author_2": " Tack A",
          "author_3": " Ehlke M",
          "author_4": " Zachow S",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Ambellan (2018)",
          "city_last_author": "Berlin",
          "country_last_author": "Germany",
          "journal": "International conference on\nMedical Imaging with Deep Learning",
          "latitude": 52.517037,
          "latitude_random": 53.28294583094045,
          "link_to_pubmed": null,
          "longitude": 13.388860000000001,
          "longitude_random": 14.294300530969585,
          "title": "Automated Segmentation of Knee Bone and Cartilage combining Statistical Shape Knowledge and Convolutional Neural Networks Data from the Osteoarthritis Initiative",
          "year": 2018
         },
         {
          "affiliation_last_author": "Zuse Institute Berlin (ZIB), Medical Planning Group",
          "algorithm_type": "hybrid",
          "author6": " Zachow S",
          "author_1": "Seim H",
          "author_2": " Kainmueller D",
          "author_3": " Lamecker H",
          "author_4": " Bindernagel M",
          "author_5": " Malinowski J",
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Seim (2010)",
          "city_last_author": "Berlin",
          "country_last_author": "Germany",
          "journal": "﻿Medical Image Analysis for the Clinic: A Grand Challenge, Beijing.",
          "latitude": 52.517037,
          "latitude_random": 53.07193170983046,
          "link_to_pubmed": null,
          "longitude": 13.388860000000001,
          "longitude_random": 15.158076554539178,
          "title": "﻿Model-based auto-segmentation of knee bones and cartilage in MRI data",
          "year": 2010
         },
         {
          "affiliation_last_author": "Biomediq and University of Copenhagen",
          "algorithm_type": "hybrid",
          "author6": null,
          "author_1": "Dam ﻿E B",
          "author_2": " Lillholm M",
          "author_3": " Marques J",
          "author_4": " Nielsen M",
          "author_5": null,
          "author_7": null,
          "author_8": null,
          "bibtex_id": "Dam (2015)",
          "city_last_author": "Copenhagen",
          "country_last_author": "Denmark",
          "journal": "﻿SPIE  Journal of Medical Imaging, vol. 2, pp. 024001-024001",
          "latitude": 55.686724,
          "latitude_random": 56.65665214388912,
          "link_to_pubmed": "https://www.ncbi.nlm.nih.gov/pubmed/26158096",
          "longitude": 12.570072,
          "longitude_random": 14.43693755775501,
          "title": "Automatic segmentation of high- and low-field knee MRIs using knee image quantification with data from the osteoarthritis initiative",
          "year": 2015
         }
        ]
       },
       "layer": [
        {
         "data": {
          "format": {
           "feature": "countries",
           "type": "topojson"
          },
          "url": "https://vega.github.io/vega-datasets/data/world-110m.json"
         },
         "height": 750,
         "mark": {
          "fill": "white",
          "stroke": "lightgray",
          "strokeWidth": 1.5,
          "type": "geoshape"
         },
         "projection": {
          "type": "equirectangular"
         },
         "width": 1250
        },
        {
         "data": {
          "name": "data-861c1f453e34c20f068f63b2a7d2c18f"
         },
         "encoding": {
          "color": {
           "field": "algorithm_type",
           "type": "nominal"
          },
          "latitude": {
           "field": "latitude_random",
           "type": "quantitative"
          },
          "longitude": {
           "field": "longitude_random",
           "type": "quantitative"
          },
          "size": {
           "value": 100
          },
          "tooltip": {
           "field": "bibtex_id",
           "type": "nominal"
          }
         },
         "mark": "circle"
        }
       ]
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 3 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import coordinates of countries to create the background map\n",
    "countries = alt.topo_feature(data.world_110m.url, 'countries')\n",
    "\n",
    "# create map\n",
    "background = alt.Chart(countries).mark_geoshape(\n",
    "    fill        = 'white',\n",
    "    stroke      = 'lightgray',\n",
    "    strokeWidth = 1.5\n",
    ").project(\n",
    "    \"equirectangular\"\n",
    ").properties(\n",
    "    width  = 1250,\n",
    "    height = 750\n",
    ")\n",
    "\n",
    "# create points\n",
    "points = alt.Chart(literature).mark_circle().encode(\n",
    "    longitude = 'longitude_random:Q',\n",
    "    latitude  = 'latitude_random:Q',\n",
    "    size      = alt.value(100),\n",
    "    color     = 'algorithm_type',\n",
    "    tooltip   = 'bibtex_id' # name of each point when hovering\n",
    "    \n",
    ")\n",
    "\n",
    "# show\n",
    "background + points "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a name=\"dependencies\"></a>\n",
    "## 4. Print dependences"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dependences are fundamental to record the **computational environment**.   \n",
    "\n",
    "- Use [watermark](https://github.com/rasbt/watermark) to print version of python, ipython, and packages, and characteristics of the computer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPython 3.7.4\n",
      "IPython 7.8.0\n",
      "\n",
      "wget 3.2\n",
      "pandas 0.25.1\n",
      "numpy 1.17.2\n",
      "geopy 1.20.0\n",
      "altair 3.2.0\n",
      "vega 2.6.0\n",
      "vega_datasets 0.7.0\n",
      "watermark 2.0.1\n",
      "\n",
      "compiler   : Clang 4.0.1 (tags/RELEASE_401/final)\n",
      "system     : Darwin\n",
      "release    : 17.7.0\n",
      "machine    : x86_64\n",
      "processor  : i386\n",
      "CPU cores  : 4\n",
      "interpreter: 64bit\n",
      " \n",
      "last updated: Tue Nov 26 2019 09:35:00 CET\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "\n",
    "# python, ipython, packages, and machine characteristics\n",
    "%watermark -v -m -p wget,pandas,numpy,geopy,altair,vega,vega_datasets,watermark \n",
    "\n",
    "# date\n",
    "print (\" \")\n",
    "%watermark -u -n -t -z "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}