Florida researchers use big data to predict Covid-19 spread
July 28, 2020
The National Science Foundation (NSF) is funding researchers at Florida Atlantic University to develop a model of Covid-19 spread using big data analytics.
Public health efforts depend heavily on predicting how diseases such as Covid-19 spread across the globe. Researchers from Florida Atlantic University’s College of Engineering & Computer Science in collaboration with LexisNexis Risk, a data technology and analytics company, have received the NSF Rapid research grant and will leverages experience in modelling ebola spread.
Researchers will use big data analytics techniques to develop computational models to predict the spread of the disease using forward simulation from a given patient and the propagation of the infection into the community; and backward simulation tracing a number of verified infections to a possible patient zero. Users of the models and algorithms developed by FAU and LexisNexis Risk will conform to applicable requirements of HIPAA and other privacy regulations.
The project will also provide quick and automatic contact tracing and is expected to help reduce the number of patients infected with Covid-19 and virus-related deaths. This methodology, which includes coalition-building efforts, will also support a range of other public health issues.
“This National Science Foundation grant will enable our researchers to advance knowledge within the field of big data analytics as well as across different fields including medical, health care and public applications,” said Stella Batalama, dean of the College of Engineering & Computer Science. “Through our collaboration with LexisNexis Risk, we will jointly address public health concerns of national and global significance using cutting-edge computer science, big data analytics, data visualisation techniques and decision support systems.”
Big data are changing how models are used to understand the dynamics of disease propagation. The FAU project, led by Borko Furht (pictured), a professor in the department of computer and electrical engineering and computer science, will use a risk score approach in modelling and predicting Covid-19 spread.
“The HPCC systems team at LexisNexis Risk has an outstanding relationship with Furht and FAU,” said Flavio Villanustre, vice president of Georgia, USA, based LexisNexis Risk. “FAU and LexisNexis Risk have been collaborating on several projects over the last five years. Our most recent work involved the NSF grant for modelling ebola using the HPCC systems platform and big data analytics. We are grateful to the NSF, FAU and Furht for their continued investment in research that helps the community.”
For the project, Covid-19 spread patterns will be fed into a decision support system (DSS), which also contains information about social groups or individual people. Social groups could include nurses and doctors who had contact with a patient infected with Covid-19, passengers who travelled on the same plane with an individual diagnosed with Covid-19, or family members living with someone who contracted Covid-19. Based on spread patterns, the DSS will then calculate probabilities for a social group or a given person to become infected with Covid-19. Data will be provided as reports to appropriate state and government agencies so they can immediately contact and test people who have a high score related to the person who is infected with Covid-19.
“The data analytics expertise we will receive from LexisNexis Risk will enable us to develop a model that will automatically and quickly identify every contact of an infected person,” said Furht. “Our approach will be much faster and more efficient than methods that are done manually and we expect it to significantly reduce infection rates and the number of deaths in the USA and around the world.”
Members of the FAU team include: Taghi Khoshgoftaar, Motorola Professor; Waseem Asghar, associate professor; Ankur Agarwal, a professor; Behnaz Ghoraani, an associate professor and a fellow of FAU’s Institute for Sensing & Embedded Network Systems Engineering; and Mirjana Pavlovic, an instructor. They all work in FAU’s department of computer and electrical engineering and computer science.
The LexisNexis Risk team includes Villanustre, senior director Arjuna Chala, senior architect Roger Dev, and Jesse Shaw, principal statistical modeller.
“Because of a lack of actual social network data, mathematical compartmental modelling has been restricted to hypothetical populations,” said Furht. “However, emerging LexisNexis Risk technologies could accelerate the accumulation of knowledge around disease propagation in the USA. For our research, we plan to calculate various scores related to Covid-19 spread including population density rank, household mortality risk, street level mortality risk and county mortality risk.”