University of Maryland Researchers Awarded NSF RAPID Grants to Bolster COVID-19 Response

University of Maryland researchers have been awarded National Science Foundation (NSF) RAPID grants to address the current COVID-19 crisis.

NSF is working closely with the scientific research community to bolster the national response to COVID-19. The agency is funding dozens of research projects on COVID-19 to mobilize the scientific community to better understand and develop measures to respond to the virus.  NSF issued a letter to researchers inviting proposals for rapid response research grants related to the virus to help inform and educate the public about virus transmission and prevention, and develop effective strategies for addressing this challenge at the local, state and national levels.  Support for these efforts is made through NSF's Rapid Response Research (RAPID) funding mechanism, which enables the agency to quickly process and support research that addresses an urgent need.

Learn more about the University of Maryland recipients of NSF RAPID grants below:

Using Location-based Big-Data to Model People's Mobility Patterns During the COVID-19 Outbreak

Kathleen Stewart (Principal Investigator)
Debbie Niemeier (Co-Principal Investigator)
Junchuan Fan (Co-Principal Investigator)

The outbreak of COVID-19 in the U.S. provides an important opportunity for researchers to study the impacts of a rapidly expanding pandemic on human mobility. This research investigates how to measure changes in collective movement of people in response to the fast-evolving COVID-19 outbreak using large datasets of passively collected location data. It examines how locations within a state respond to public policy implementation and times of critical public messaging. Detailed knowledge on movement patterns of people can help public officials identify hotspots and critically isolated populations, as well as shed light on those groups who continue to travel for work or other purposes. This research contributes to improving the public response to an emergency and contributes to bridging different stakeholder mitigation strategies.

Detailed knowledge of how people respond to a fast-spreading global pandemic is very limited and our understanding of these responses is mostly for small areas. This research will use a near real-time location-based dataset passively collected through the use of location-based apps during the period of pandemic. The project will develop scalable, big location-based algorithms to extract trips and examine the evolution of mobility patterns throughout the pandemic, and identify different mobility patterns. The research team will develop map-reduce based distributed algorithms to scale up mobility measure calculations based on the big location-based data as well as develop entropy measures to capture the time-varying characteristics associated with the travel patterns, and design strategies to correct biases that may be present in the location data. The methods and results of this research will be useful for understanding mobility during other hazards that affect communities, such as severe flooding to understand how travel is changed as a result of imperatives stemming from both the hazard and policy directives.

Combining Big Data in Transportation with Hospital Health Data to Build Realistic "Flattening the Curves" Models during the COVID-19 Outbreak

Deb Niemeier (Principal Investigator)
Kartik Kaushik (Co-Principal Investigator)

The outbreak of COVID-19 in the U.S. provides an important opportunity for researchers to improve flattening curve models which can be used to assess and even spatially optimize health care during a rapidly expanding pandemic. This Rapid Response Research (RAPID) project will take advantage of the large-scale availability of location-sensing devices and apps that produce big data on mobility patterns that can be used to better optimize the use of healthcare facilities. This research brings together rapidly unfolding health data with real-time data on mobility. The researchers will examine how these two critical data resources can be linked to better inform policy, identify emerging hotspots, and target critical actions during a pandemic. This research will help public officials to better understand and adapt to changing conditions as a health emergency arises and expands.

The spread of the “flattening curves” graphic was significant in promoting public understanding of the criticality of social distancing. These curves, however, were based on simulated data. This research will collect and examine mobility data and public health data to model flattening curves using real data. The researchers will combine big data from location-based apps and cellphones with Electronic Medical records from UMMS hospitals, including data on COVID-19 tests, and patient demographics and prognostics. New modeling approaches that quantitatively measure change in collective movement behaviors in response to the fast-evolving COVID-19 outbreak will be linked to hospital usage and capacity. The methods of this research will extend our knowledge of highly integrated systems, like transportation and health, and better prepare the public for future disasters.

Advanced Topic Modeling Methods to Analyze Text Responses in COVID-19 Survey Data

Philip Resnik (Principal Investigator)

As the COVID-19 pandemic continues, public and private organizations are deploying surveys to inform responses and policy choices. Survey designs using multiple choice responses are by far the most common -- "open ended" questions, where survey participants provide a longer-form written response, are used far less. This is true despite the fact that when you allow people to provide unconstrained spoken or text responses, it is possible to obtain richer, fine-grained information clarifying the other responses, as well as useful "bottom up" information that the survey designers did not know to ask for. A key problem is that analyzing the unstructured language in open-ended responses is a labor-intensive process, creating obstacles to using them especially when speedy analysis is needed and resources are limited. Computational methods can help, but they often fail to provide coherent, interpretable categories, or they can fail to do a good job connecting the text in the survey with the closed-end responses. This project will develop new computational methods for fast and effective analysis of survey data that includes text responses, and it will apply these methods to support organizations doing high-impact survey work related to COVID-19 response. This will improve these organizations' ability to understand and mitigate the impact of the COVID-19 pandemic.

This project's technical approach builds on recent techniques bringing together deep learning and Bayesian topic models. Several key technical innovations will be introduced that are specifically geared toward improving the quality of information available in surveys that include both closed- and open-ended responses. A common element in these approaches is the extension of methods commonly used in supervised learning settings, such as task-based fine-tuning of embeddings and knowledge distillation, to unsupervised topic modeling, with a specific focus on producing diverse, human-interpretable topic categories that are well aligned with discrete attributes such as demographic characteristics, closed-end responses, and experimental condition. Project activities include assisting in the analysis of organizations' survey data, conducting independent surveys aligned with their needs to obtain additional relevant data, and the public release of a clean, easy to use computational toolkit facilitating more widespread adoption of these new methods.

Assessing the Social Consequences of COVID-19

Long Doan (Principal Investigator)
Jessica Fish (Co-Principal Investigator)
Liana Sayer (Co-Principal Investigator)

This project examines the impacts of COVID-19 and states' and local governments' social distancing directives on behavior, time spent with others, use of technology, and mental and physical wellbeing. The objective of the project is to investigate these daily life impacts in real time and to analyze how these impacts are affected by sociodemographic characteristics that affect time use and well-being. Data are leveraged from several hundred respondents? daily time use before the pandemic along with data collected during and after the pandemic to create a natural experiment that isolates the effects of the pandemic on changes in behavior. Among the products of this research are evidence-based recommendations to address the social consequences of the pandemic.

This project collects data for the second and third waves of a three-wave panel study, the second wave during the pandemic with shelter-at-home and lockdown orders in place and the third wave after the pandemic has subsided and orders have been relaxed. Data for these two waves consist of survey responses and 24-hour time diaries collected from 2,000 respondents from online crowdsourcing platforms. This sample includes a smaller sample from whom data were collected before the pandemic. Data are collected on sociodemographics, typical sleep, work, and exercise patterns, and arrangements for housework and carework to investigate effects on time use and wellbeing.

Energy-Efficient Disinfection of Viral Bioaerosols in Public Spaces: Vital for Lifting of the ?Stay-at-Home? Orders during the Covid-19 Outbreak

Jelena Srebric (Principal Investigator)

This project will provide an analytical framework to assess potential reduction of infection risks from COVID-19 viral bioaerosols in public spaces, including school buses, classrooms, and retail stores. Viral bioaerosols may cause infection for occupants staying both near and far away from infected people, whether staying indoors at the same time or not. Upper-room germicidal ultraviolet (UR-GUV) light can provide a real-time air disinfection solution with a relatively small energy footprint if its light effectively interacts with the bioaerosol both in the air and on surfaces. This project will develop and disseminate an open-source numerical analytical framework including assessment of UR-GUV disinfection and make it publicly available online to provide a free resource useful for helping to control the spread of airborne COVID-19 infections in public spaces. An effective, real-time, and sustainable engineering solution for air indoor space disinfection is an important precaution to help prevent the spread of COVID-19, particularly in the context of efforts to restart the nation's economy.

The project will develop numerical methods based on Computational Fluid Dynamics (CFD) to reproduce the processes for viral bioaerosols spread by indoor airflow, removed by exhaust, inactivated by UR-GUV, inhaled by the occupants, and deposited onto surfaces in public spaces of varied spatial scales, ventilation systems, as well as population size and density. This project will also optimize the application of ceiling fans to improve UR-GUV disinfection efficacy. The investigation will provide new insight on infection risk due to viral aerosols and infection control by UR-GUV for surfaces contaminated by viral bioaerosols. In addition, the project will consider two UV-C sources, one by traditional mercury vapor UV-C lamps (UV-C-MV) and another by UV-C-LED for their energy efficiency. The comparison of the two UV-C sources in terms of disinfection, energy efficiencies, and operation cost holds promise for a sustainable UR-GUV solution for minimizing infection risk in public spaces.

May 10, 2020