COVID-19 spatiotemporal research with workflow-based data analysis

Given the pertinence and acceleration of the spread of COVID-19, there is an increased need for the replicability of data models to verify the veracity of models and visualize important data. Most of these visualizations lack reproducibility, credibility, or accuracy, and are static, which makes it difficult to analyze the spread over time. Furthermore, most current visualizations depicting the spread of COVID-19 are at a global or country level, meaning there is a dearth of regional analysis within a country. Keeping these issues in mind, a replicable, efficient, and simple method to generate regional COVID-19 visualizations mapped with time was created by using the KNIME software, an open-source data analytics platform that can create user-friendly applications or workflows. For this analysis, Albania, Sweden, Ukraine, Denmark, Russia, India, and Australia were closely observed. Among the maps generated for the aforementioned countries, it was noticed that regions with a high population or high population density were often the epicenters within their respective country. The regions caused the virus to spread to their neighboring regions: kickstarting the “domino effect”, leading to the infection of another region until the country is overwhelmed with cases—what we call a proximity trend. These dynamic maps are crucial to fighting the pandemic because they can provide insight as to how COVID-19 spreads by providing researchers or officials with an accurate and insightful tool to aid their analysis. By being able to visualize the spread, health and government officials can dive deeper to identify the sources of transmission and attempt to stop or reverse them accordingly.