ISSS608 2016-17T3 Group15 Proposal

ISSS608 Visual Analytics and Applications

Group 15 Project
Visualisation with R:
Characterisation of Pandemic Spread

Proposal

Poster

Application

Report

Project Title: Characterisation of Pandemic Spread

Prepared by Group 15

Team Members:

• Chua Gim Hong
• Huang Liwei
• Ngo Siew Hui

Project Description

This project is based on VAST Challenge 2010 – Characterisation of Pandemic Spread, which involves the analysis of hospitalisation records relating to a major pandemic spread across the world in 2009. With the use of R, the project aims to develop a visualisation tool to analyse the illness across these countries, so as to help characterise the spread of the disease.

Background

There was a major disease outbreak that spanned several cities across the world in 2009. Such diseases tend to spread fast and are fairly difficult to combat. Hence, health officials are seeking for visualisation tools to analyse the hospitalisation records across these countries, so as to help characterise the spread of the disease.

Data

Note: The datasets used for this challenge are synthetic, with a blend of computer and human-generated data. No external data is needed to perform the analysis as all information necessary to form working hypotheses are provided in the datasets.

The datasets contained hospital admittance and death records for eleven cities involved in the pandemic, namely:

• Aleppo
• Colombia
• Iran
• Karachi
• Lebanon
• Nairobi
• Saudi Arabia
• Thailand
• Turkey
• Venezuela
• Yemen

Ojectives

The project aims to develop a visualisation tool using R programming to perform the following analysis:

• Characterise the spread of the disease, taking into consideration symptoms of the disease, mortality rates, temporal patterns of the onset, peak and recovery of the disease.
• Compare the outbreak across cities, including the timing of outbreaks, numbers of people infected and recovery ability of the individual cities.
• Identify anomalies from the hospitalisation records, if any.

Approach

• [placeholder]
• [placeholder]

Motivations

Through this project, we hope that the visualisation tool developed can help health officials analyse hospitalisation records for the next disease outbreak. With the help of visual analytics, the tool aims to save them analysis time so that they can react quickly to the pandemic spread.

This would be particularly relevant in recent years as there have been widespread disease outbreaks leading to an alarming number of cases in the affected regions. For example, the Ebola outbreak has led to major loss of lives (~11K), and there is growing evidence that the Zika outbreak would lead to high risk of birth defects and other neurological disorders. For both outbreaks, there have been significant socioeconomic disruptions for the affected regions.

• Ebola Outbreak (2014 - 2016)
• Zika Outbreak (2015 - 2016)

Challenges

How to create visualisations using R programming:

• Steep learning curve to pick up R programming skillsets

Lack of domain knowledge in hospitalisation records:

• Longer time required for data exploration and cleaning

Lack of domain knowledge in epidemic and pandemic outbreaks:

• More background research is required in order to design more insightful visualisations

Milestones & Expected Outcome

• [placeholder]
• [placeholder]

References

• VAST Challenge 2010 – Characterisation of Pandemic Spread
• 2014-2016 Ebola Outbreak in West Africa
• CDC Zika Virus
• R for Data Science
• Tidyverse R Packages
• R Shiny Tutorial