COVID-19: How cell phones are helping to track future cases

Why cell phones can hold the key to tracking future cases of COVID-19 with artificial intelligence.

COVID-19: How cell phones are helping to track future cases

TechRepublic's Karen Roby talked with Ari Trachtenberg, a professor of electrical and computer engineering at Boston University, about the work researchers are doing to help in the fight against COVID-19. The following is an edited transcript of their interview.

Ari Trachtenberg: This started with me staying at home and trying to figure out what we could possibly do to help with the COVID-19 effort. And we realized that after we plateau and we manage to get a COVID-19 under control, we'll still need to be able to make sure that it doesn't start spreading again. And one way to do that is to make sure we isolate people who are sick as quickly as possible and prevent them from exposing the rest of the community. We're thinking of how to do this, and the natural solution had to do with cell phones, which are always around us and have ways to communicate with each other at short ranges.

SEE: Coronavirus: Critical IT policies and tools every business needs (TechRepublic Premium)

Karen Roby: Talk about the technology and privacy concerns, too.

Ari Trachtenberg: That's exactly right. We are very sensitive to the fact that we already have a lot of information out there in the digital space, and we're worried about that potentially being misused for other purposes. Our approach just has your phone, has an app on your phone, that just sends out random messages, completely random. Has no connection to your information, your location, your name, anything. And those random chirps, they're called, can be used to then identify later whether you were around someone who ended up being sick and that that person when they get sick will report their own random numbers that they have submitted. And those random numbers, when I look and correlate them with random numbers that I've heard in my area, will tell me maybe I need to go get a test.

Karen Roby: And so from there, Ari, what will be the next step once you've compiled that information, how does it then move down the line so that people will know what to do?

Ari Trachtenberg: The idea is when you get sick, you go to your doctor or to get a test. If the test comes back positive, then you will on your own give the doctor all the numbers that you have been broadcasting around as you've been walking with your phone. The doctor will upload those to a central registry and everyone else who's walking around will then look at the registry and see whether they heard your numbers anywhere. And if they did, they know that they might be sick and they need to go get a test, and then this process perpetuates itself. They go get a test if they ended up being positive, that goes up to the registry, etc.

SEE: Cybersecurity: Let's get tactical (free PDF) (TechRepublic)

Karen Roby: Talk a little more about your work.

Ari Trachtenberg: I work in cybersecurity, and in fact this is a slight tweak of existing attacks. I work on attacks and defenses, ways that malicious entities try to identify your location. And so we were aware of that and tried to design a system that was resistant to that so that you cannot be tracked, so that it cannot be determined where you are. I want to stress that the location information that you have on your phone where you are at all times is extremely personal information. It says things like where you slept last night, what kind of medical conditions you're being treated for, depending on where you go. Who your friends are, what your political and religious interests are. This is very, very deeply personal information that needs to be guarded and we believe that this solution provides you the opportunity to protect your privacy while also sharing in helping others and yourself not get this disease.

Karen Roby: Who are you collaborating with on this effort?

Ari Trachtenberg: The original work was a collaboration with professor Ran Canetti in computer science and professor Mayank Varia also in computer science, also part of the Hariri Institute and part of the risk center. And this was a BU effort. We reached out to various other colleagues in BU. In fact, we initially discussed it with them, and now we're actually part of a greater initiative that's led by (Massachusetts Institute of Technology) MIT but connected to a variety of other institutions. All of us are working together trying to get this implemented as fast as possible.

Karen Roby: How concerned are you at the moment with cybersecurity?

Ari Trachtenberg: I am very concerned with cybersecurity. Generally, malicious attackers look for situations like this where there's widespread confusion and rules and regulations get to be broken or bent for the greater good. I'm also concerned with mission creep in the sense that it's at times like this when we expect our governments to go through extreme measures to protect us, but those extreme measures sometimes linger on even after the extreme situation has gone away. I think it's important to be aware of that upfront while designing a system like this to think about, making sure that it cannot be taken advantage of later on down the line, not necessarily by the government, but even by other people who just see an opportunity.

Also see

20200408-bostonuni-karen.jpg

Karen Roby talks with Ari Trachtenberg of Boston University about an app he is creating to track COVID-19 without compromising privacy. 

Image: Mackenzie Burke