Reliable Analytics for Disease Indicators: Leveraging Smart Devices to Predict Health; An Actor Network Theory Approach to the Improvement of Health Care in Rural America
Moens, Charles, School of Engineering and Applied Science, University of Virginia
Gorman, Michael, University of Virginia
Barnes, Laura, EN-Eng Sys and Environment, University of Virginia
Boukhechba, Mehdi, EN-Eng Sys and Environment, University of Virginia
Rural Americans’ health is notably worse than their urban counterparts, largely due to structural and economic limitations that previously could not be circumvented. However, with the advent of advanced mobile sensing technology, rural Americans can be notified of potential health concerns simply by having a smartphone on their person. By essentially circumventing inadequate and understaffed health care facilities, I posit that disease prediction technology can effectively bridge the gap between urban and rural health care and improve the standard of living for some of America’s most neglected constituents.
My technical research capstone project is headed by Assistant Professor Mehdi Boukhechba and Associate Professor Laura Barnes, both of the Department of Engineering Systems and Environment. Aimed at assessing warfighter readiness for DARPA, my capstone team is focused on optimizing both battery consumption and disease prediction capabilities. Utilizing a mobile sensing app called Sensus, the team planned to conduct a study of 30 participants over the course of 3 weeks, whose data were to be used to develop and train machine learning models to selectively turn sensors on when important information is anticipated to be detected. Due to the novel Coronavirus pandemic, the full-scale study was unable to be run. Instead, data collected from 10 capstone team members was used. In order to properly identify patterns and key metrics in real time, a customized time-series database was created, allowing for much higher data throughput than before. The technical report goes further into detail regarding study design, methods, results, and conclusions of our research.
My Science, Technology, and Society Thesis targets the discrepancy of care and consequential poor health experienced by Americans living in rural areas. It presents the descriptive situation of the average rural health care system: a population with poorer health and a lesser ability to seek out care compared with their urban counterparts. I assert that there is a reinforcing loop based on a number of factors that uniquely worsens the health of rural Americans. In order to break this loop and improve rural health care in America, a combination of government policy and technology is needed to overcome the sizable structural hurdles to quality care. Thanks to recent advances in modern wearable health devices and smartphones as well as their penetration in rural communities, unique opportunities for detecting and reporting disease have presented themselves. Previous attempts at solving this problem with technology have failed because of both 1) prohibitive costs for dedicated telemedicine equipment and 2) lack of coordination between insurance companies, medical professionals, and the government, among others. My thesis examines two states of rural health care in America: the descriptive, which presents the current state of rural health and its shortcomings, and the normative, which presents the optimal state of rural health vis-a-vis the broad implementation of wearable technology and coordinating systems. Underpinning my analysis is Katie Rodger’s definition and approach to the Actor-Network Theory framework. Using this framework, I analyze both the descriptive and normative scenarios as functions of their actors and the linkages between them. The normative scenario contains additional actors and linkages that stabilize the system’s reinforcing loops and reveal a sustainable roadmap to improving rural health. Of course, the infrastructural changes necessary to implement are sizable and require regulatory planning on behalf of the government. Thus, the second facet of my thesis utilizes the Anticipatory Governance framework in order to be prepared for the changing health care landscape. As health care data is strongly sensitive, it must be regulated and large government controls are necessary to prevent corporations providing telehealth technology from abusing their involvement in consumers’ personal matters. Additionally, the third and final framework that enables wearable technology to be trusted in consumers’ eyes is the notion of contractarian ethics. I examine how this form of ethics is the best choice to ensure both consumers and companies are in agreement with how data is being used, stored, and handled. In summary, this approach provides a comprehensive option for improving the state of rural health care in America by leveraging technology and minimizing costs.
The link between my technical capstone project and my STS thesis is clear: if disease prediction technology from my research is ever commercialized, then it will serve to amplify the power of the data collected by wearables, providing doctors and first responders with early insights not possible before. With robust early detection of illness, rural communities are able to know when they need to seek care with urgency.
BS (Bachelor of Science)
health care, rural america, wearable technology
School of Engineering and Applied Science
Bachelor of Science in Systems and Information Engineering
Technical Advisor: Laura Barnes, Mehdi Boukhechba
STS Advisor: Michael Gorman
Technical Team Members: Shalin Shah, Cameron Fard, Ian Tucker, Tucker Wilson, Hannah Katinas, Lauren Perry, Erin Barrett, Blake Ruddy
All rights reserved (no additional license for public reuse)