As society looks to mitigate the effects of COVID-19, contact tracing applications have emerged as one way to determine who around you is healthy or affected with the virus. While these apps could seemingly offer a way to avoid infection, they also raise questions about privacy and the effectiveness of the approach.
Ohio State College of Engineering researchers Ness Shroff and Dong Xuan have been developing a contact tracing app that protects your privacy. They share insight into these applications and how they might help.
Contact tracing applications allow users to determine whether they have been in contact with someone who has tested positive for COVID-19. They rely on users providing information or having their health care providers share information on their “state” (well, sick, etc.). This information could then be used to conduct efficient testing, help in understanding the scope of the current epidemic, estimate resources needed by hospitals and more.
The goal of these apps is to determine whether individuals have been in close proximity — six feet — with each other. There are different technologies that have been proposed for doing this. Most approaches use Bluetooth in their mobile phones.
Our approach uses inaudible ultrasound to determine the proximity of the users. We believe that while this technology is more difficult to implement, its key advantage is that it avoids unnecessary contacts — those greater than six feet away — from being recorded.
It depends on the app. The recent Apple and Google proposals appear to have taken security and privacy seriously, but researchers already are asking privacy and security questions. I cannot speak for all the apps out there, but users should be careful to read the agreement, especially for contact tracing apps. For our technology, users will be completely anonymized.
The goal is to allow the user to be in control of the level of information that she or he is willing to share.
For example, we will design an app so that a user can specify a range of personal information (from none to all) such as GPS information of each encounter, their age range, medical history, gender and more.
We also will work with privacy and security experts to make sure the app is secure.
While the specific app has been developed with a focus on the COVID-19 crisis, other potential applications of the app include advertising, safety notification, social networking, gaming and more.
We expect that it is critical to design the back-end (database) that would allow various apps and technologies to feed in their information in order to develop a more accurate contact graph.
You are correct.
To a large extent, the success of the tools depends on citizens’ willingness to participate. We will publicize the utility of the app in keeping the community safe.
We will also work with our local health departments and Ohio State to help make public safety announcements one of the benefits of using such an app.
Carriers do not need to participate in this effort, unless they want to donate equipment to help.
However, anonymized coarser level data collected by carriers could also be incorporated in the back end (database) to further help refine contact tracing.
Our experiments are going surprisingly well and we are cautiously optimistic.