Shiny Framework Based Visualization and Analytics Tool for Middle East Respiratory Syndrome
- 268 Downloads
People in the Middle East have been affected by the Middle East Respiratory Syndrome CoronaVirus (MERS Co-V) since 2012. New cases are continuously reported especially in the Kingdom of Saudi Arabia, and the risk of exposure remains an issue. Data visualization plays a vital role in effective analysis of the data. In this paper, we introduce an interactive visualization application for MERS data collected from the Control and Command Centre, Ministry of Health website of Saudi Arabia. The data corresponding to the period from January 1, 2019 to February 28, 2019 was used in the present work. The attributes considered include gender, age, date of reporting, city, region, camel contact, description and status of the patient. The visualization tool has been developed using Shiny framework of R programming language. The application presents information in the form of interactive plots, maps and tables. The salient feature of the tool is that users can view and download data corresponding to the period of their choice. This tool can help decision makers in the detailed analysis of data and hence devise measures to prevent the spread of the disease.
KeywordsMERS Data visualization Shiny framework Coronavirus Location-based analysis
MERS is a viral disease that was first discovered in 2012 when a patient in Saudi Arabia was diagnosed with critical respiratory distress and kidney trouble . The infection can either be asymptomatic or show symptoms such as cough, fever etc. along with difficulty in breathing . The disease has been brought under control in all Middle Eastern countries except Saudi Arabia where each month, new cases are still reported. Studies have confirmed that this disease is zoonotic and camels are its significant reservoir [3, 4]. People in close contact with infected camels and health care professionals who care for infected patients have a high risk of acquiring the infection. Many cases of community and household acquired infections have been reported from Saudi Arabia . The Ministry of Health, Saudi Arabia has issued guidelines for infection prevention, control and management.
To the best of our knowledge, there is a lack of interactive visualization tools for MERS data visualization in Saudi Arabia. This work deals with developing an application where users can interactively view information about the infection in the form of plots, tables and maps. The data used in this study was obtained from the Saudi Ministry of Health website and it includes the cases reported in Saudi Arabia from January 1, 2019 to February 28, 2019. By viewing the data visualizations, users can analyze MERS cases better, find trends, monitor the disease and help authorities set detection and prevention guidelines.
2 Data and Methods
2.1 Data Description
Description of data used.
Date of reporting
Male or female
Age of patient
Region corresponding to patient city
Yes or no
Primary or secondary
Community acquired, household contact, healthcare acquired or unknown
Dead, recovered or hospitalized
2.2 Shiny Framework
2.3 Leaflet Package
2.4 Googleviz Package
The googleviz package in R facilitates the use of Google Chart APIs . Interactive charts can be incorporated into web pages using Google Charts. The data stored in R data frames can be visualized in the form of Google Charts without uploading the information to Google. An internet connection is required to view the output rendered by this package.
3 Results and Discussion
The MERS data visualization tool was developed using R programming language. The tool consists of three sections (tabs) namely “Different Cases Analysis”, “Miscellaneous Analysis” and “Summary” section. The users can choose the period for which they would like to visualize the MERS data.
3.1 Different Cases Analysis
In the case of different cases analysis, the user can view the information as pie charts and maps, or tables. The users can view details about either all the cases together or recovered cases, death cases or hospitalized cases separately. The details regarding the cases can be viewed with respect to all cities, all regions or cities within a region. In the case of the option “cities within a region”, the user has to choose a region from the drop down list provided. Unlike conventional pie charts, the pie charts created using googleviz package are interactive in nature. By pointing the cursor over a portion in the pie chart, one can understand the actual number of cases in that particular part of the chart. In the map, places are marked based on longitude and latitude of that place. On clicking the marker displayed in the map, the name and number of cases in that place will be displayed as a pop-up. Based on the number of cases, markers are assigned colors such as red, orange or yellow. The colors red, orange and yellow represent “large number of cases”, “moderate number of cases” and “few number of cases” respectively.
3.2 Miscellaneous Analysis
The miscellaneous analysis consists of analysis based on age, camel contact and months. The users can analyze the data for all cases, death cases or recovered cases. Depending on the user’s choice, the analysis corresponding to all cities, all regions or cities within a region are displayed.
Analysis Based on Month-Wise Comparison. Month-wise comparison of the data can be carried out for “all cases”, “death cases” and “recovery cases”. The stacked bar chart for cases reported in different months is plotted based on the location type specified by the user. When the cursor is moved to a region in the bar chart, the number of cases in the particular month will be displayed corresponding to the location. The screenshot corresponding to month-wise comparison is shown in Fig. 6. The figure portrays the analysis corresponding to death cases reported in various regions. The maximum number of death cases were reported in Riyadh region during February. Monthwise analysis is useful in identifying whether the infection is related to climate.
3.3 Summary-Based Visualization
Summary based visualization gives a graphical summary of the count of the “status of the patients” for different attributes. The information is depicted in the form of stacked bar charts where the charts are plotted based on the frequency of status of patients corresponding to different values of the attributes. This visualization will help the users in getting an overall idea regarding the distribution of data with regard to the status of the patients. Figures 7 and 8 depict the screenshots of the visual summary of data for the first two months of the year 2019.
3.4 Benefits of the Application
The users can view information corresponding to their period of interest. The interactive feature of the pie chart prevents the chart from being cluttered with description. The use of maps to represent the information will give an idea regarding the spatial distribution of MERS cases. These maps will help the health authorities to identify the areas with large number of cases and hence alert the hospitals and the general public regarding that. The interactive nature of the table helps the users in analyzing the data as per their requirement. A salient feature of the application is that the users can easily download details corresponding to the period of their choice. Analyzing the data based on camel contact will aid health authorities to identify the areas where many such cases are present and hence intensify the awareness programs to reduce the rate of infection.
In this paper, we have created an interactive visualization tool for MERS Co-V infection cases based on details of cases reported in Saudi Arabia. The attributes used include the date of reporting, city, region, age, gender, description of the disease, camel contact and status of the patient. The user can view details regarding all cases, recovered cases, death cases or hospitalized cases for all the cities, all regions or cities within a region. Our tool provides the flexibility to view information in the form of charts and maps, or downloadable tables. The analysis is also carried out based on age group, camel contact and month-wise cases. By viewing the maps, users can easily differentiate between places based on the number of cases. Moreover, the users can view a visual summary of the number of cases for different values of attributes based on the status of the patients. Understanding and analyzing the disease related information can help decision makers in setting guidelines for preventing and controlling the spread of disease. Location-based analysis of the infection is highly essential in formulating region specific awareness programs to reduce the rate of infection.
- 2.John, M., Shaiba, H.: Main factors influencing recovery in MERS Co-V patients using machine learning (2019). https://doi.org/10.1016/j.jiph.2019.03.020
- 6.CRAN-Package shiny. https://CRAN.R-project.org/package=shiny
- 7.CRAN-Package leaflet. https://CRAN.R-project.org/package=leaflet
- 8.CRAN-Package googleVis. https://cran.r-project.org/package=googleVis