A review of mobile sensing in the era of COVID-19

Portable sensing has demonstrated its power in diffuse and effective monitoring of COVID-19 at various population scales and time duration, according to a study published in Health Data Sciencea Science Partner Magazine.

Behind the work are researchers in the Sensing System for Health Laboratory led by Dr. Laura Barnes at the University of Virginia. They have worked to advance health and well-being using sensor technologies and mobile data analysis.

Portable Sensor, a digital monitoring tool, takes advantage of sensors built into mobile devices such as smartphones and wearables. As mobile sensing has become a promising method for monitoring epidemic trajectories through data collection at individual, community, and global scales, this paper investigated study designs, expected health outcomes, and current limitations of such mobile-based human work to guide future practice. As such, this paper stands out among a group of articles on the use of mobile devices to respond to COVID-19.

“We reviewed 1) objectives and designs of the current work, 2) sensing duration and population coverage, 3) outcomes and limitations, in order to better categorize and understand this topic.” says Zhiyuan Wang, a doctoral student at Sensing Systems for Health Lab.

“The present work demonstrated the ability of mobile sensing to not only 1) detect infection status remotely, but also 2) track disease progression longitudinally in order to personal medicine3) to passively track exposures and 4) to monitor the impact of the pandemic on a large scale population health”Associate Professor Laura Barnes, Laboratory Director.

However, technical and societal limitations still exist, including data availability and system adoption challenges, clinical and application issues, privacy and ethical concerns. These limitations have hampered further actions by computer scientists, clinicians, and epidemiologists in leveraging mobile sensing for human health.

Existing or emerging technologies may provide a solution to these limitations. For example, advances in data analytics and machine learning methods may help improve data quality due to its ability to handle sparse, disparate, and multimodal sensor data streams. Also, mobile sensing can be performed at larger scales, particularly in clinical settings, by taking advantage of the next generation of sensors and sensing platforms.

Other stakeholders can also influence how mobile sensing achieves clinical and social benefits. These efforts can include mitigating potential threats to privacy, equity, and health inequalities; Promote technology and health literacy in all societies; And make decisions based on trust and jointly correctly balancing risks and benefits.

Barnes and her team hope to see more work as computer scientists, clinicians and epidemiologists design and implement the study in collaboration with experts in the social sciences and public policy to enable more effective, scalable, and socially equal mobile health systems for infectious diseases.


Journal reference: