Advertisement

UC Irvine researchers develop app to help track COVID-19, but will people use it?

TrackCOVID
Tyler Yasaka, a UC Irvine researcher, co-developed a contact-tracing application for smartphones that could help keep track of people exposed to COVID-19.
(Don Leach / Staff Photographer)
Share via

UC Irvine researchers have developed a phone application that could potentially help stem the spread of COVID-19 by tracking and isolating people who may have been exposed to the deadly virus.

But activist groups and legal experts are warning of potential privacy encroachments that could occur if tracking technology is widely adopted.

TrackCOVID is a proposed smartphone application for contact tracing, a practice Gov. Gavin Newsom said on Tuesday was needed before the state’s economy can reopen. Contact tracing is the process of finding out who has been in contact with a person who has recently tested positive for COVID-19.

Advertisement

Brandon Lehrich, who co-developed TrackCOVID, said contact tracing has traditionally been accomplished manually, with governments hiring hundreds of individuals to go out into the community to track down people who may have been exposed. Provoked by the rapid spread of COVID-19, institutions like Apple, Google, Stanford and the Massachusetts Institute of Technology have been developing digital methods of contact tracing.

UCI’s program uses a novel approach with quick response, or QR, codes.

“The drawback of manual contact tracing is a lot of people just don’t remember where they have been and who they have been in contact with, nor how to contact those people,” Lehrich said. “However, if you are automating the process digitally then you hope you can cast as broad a net as possible and inform as many people as possible related to how many people you came in contact with.”

The UCI tool requires the public to scan QR codes that would ideally be posted outside of high-traffic areas, like grocery stores or universities. After scanning the code with a phone camera, the application registers that the individual is at that particular place. If somebody who had been at that location eventually tests positive for COVID-19, then all other users who were registered at that location would be notified on their phone by the application that they may have been exposed to the virus.

The concept was developed for Android, but users don’t actually have to download anything to their phone. After taking a photo of a code, the TrackCOVID website will open up on a web browser and register the data.

Tyler Yasaka, the project’s lead author, said privacy was a significant consideration for the group as they developed the tool.

One of the primary ways the developers sought to preserve privacy was by having the app register people’s whereabouts on an anonymous graph. When people photograph the QR code, they are placed into an area, called a checkpoint, on a graph that isn’t tied to any geography.

“It’s just a checkpoint — just to show that these people congregated in the same area, but it doesn’t say where,” Yasaka said. “We are trying to track the virus, not people.”

The app doesn’t ask for personal information, including email or phone.

“The approach we took is to store and process as little information as possible,” Yasaka said. “You don’t have to worry about info leaking if it’s not there. As far as when you self report — that’s probably the highest area of concern — that information is not even stored on your device. It just tells the server, ‘Hey, update these checkpoints as potentially being exposed.’

“There is no concept of you on the server. It just flags those checkpoints. All the server knows is somebody over here at these checkpoints has been exposed.”

The researchers designed the application as an open-source code, which allows for others to improve it.

“Hopefully people can build on top of it and potentially merge technologies,” Lehrich said.

On Thursday, the American Civil Liberties Union published a paper on the privacy concerns posed by contact tracing technology.

“We may well see the development of a privacy-preserving, technology-augmented contact-tracing system that makes a significant contribution to fighting the COVID-19 pandemic,” the article reads. “But there are risks to fundamental civil liberties posed by poorly designed systems in this space, and a poorly-designed system may be ineffective or even make the pandemic worse. We need a sober consideration of the risks and tradeoffs of such a system so that it protects not only the fundamental right to health, but also our rights of privacy and free association.”

Local ACLU personnel were not available to discuss the specifics of the UCI app.

The Electronic Frontier Foundation, a digital-rights group, recently reported that “the risks [contact tracing] poses to individual privacy and civil liberties is considerable.”

Privacy concerns could pose a significant obstacle for widespread adoption of contact-tracing applications.

Some of the current contact-tracing technology being developed, like from MIT, uses Bluetooth to track a person’s whereabouts. Yasaka said his team chose to not use Bluetooth because the general public likely would feel uncomfortable with having Bluetooth on all day.

“It is private,” Yasaka said, “but that feeling of having it always on can make people nervous.”

The UCI team is hoping that social encouragement will cause people to take part in its contact-tracing application.

“Hopefully, it can become a social practice that people encourage you to help in this effort to contain the disease,” Yasaka said.

A significant number of the population has to take part in the contact-tracing app for any impact to be made, though what number isn’t really known. The Singapore government launched a contact-tracing application late last month. Yasaka said about 12% of the Singapore population is using the app. He believes the percentage generally needs to be closer to 50% for contact tracing to work.

“For any contact tracing solution to work, you really need high participation from everyone in the population” Lehrich said. “For this, you would have everybody essentially scanning these as they enter a public place.”

It could be difficult to get the public to buy into digital contract tracing because its efficacy has never truly been proven.

The UCI researchers said the only proof of the potential success of digital contact tracing is a University of Oxford study published March 31, which found that it could be effective with enough people using it and complying with quarantine and isolation measures.

Another hurdle is getting businesses to put the QR codes up, as well as getting politicians on board. Yasaka said the app requires very little monitoring — it largely is a peer-to-peer program — but policy makers and public health and privacy experts should be in charge of the app in conjunction with one another.

“I think this can be a really effective solution if we can get the ball rolling,” Yasaka said. “The concern is that it never takes off.”

Support our coverage by becoming a digital subscriber.

Advertisement