Privately Identify and Help Track the Spread of COVID-19 with COVID Nearby
COVID Nearby uses crowdsensing to empower people to securely generate where it is present and how it spreads
Share this on:
“New Data Platform – COVID Nearby:
Privately share your health symptoms to track COVID-19.
See how many people report symptoms near you, your friends, or your family.
Your privacy comes first.”
A team of Rutgers professionals developed and launched COVID Nearby, a data platform used to identify and track the spread of COVID-19. Users can securely and privately report their location and health symptoms, and use the interactive map to see the spread of COVID-19 in their communities.
The key distinguishing feature of COVID Nearby is its strong privacy guarantee since it uses differential privacy.
COVID Nearby guarantees privacy to all individuals using the app. According to Rutgers Business School, the app will also provide researchers with insights about the privacy preferences of individuals during health emergencies.
Rutgers Business School’s Jaideep Vaidya, who is also an IEEE Computer Society Volunteer, received funding from the National Science Foundation to build a secure and privacy-protecting data platform. With that, he led the multi-disciplinary team of Rutgers professors in the development of COVID Nearby.
The professors include scientists from Rutgers Business School, the School of Public Health, the School of Communication and Information, who have expertise in epidemiology, behavioral analysis, and data privacy. COVID Nearby is being developed by the Rutgers Institute of Data Science, Learning, and Applications with support from the National Science Foundation.
To fight COVID-19, science needs to know where it is present and how it spreads within our communities. COVID Nearby uses crowdsensing to empower people to generate this information. Even a few thousand responses can help track the spread; so all of our participation is crucial to win this fight.
As of now, the data and visualization are only provided for the USA. However, the app is accessible worldwide, and any person can contribute their data.
The underlying technology is generally applicable and is currently being explored in other countries.
COVID Nearby reports are completely anonymous – no personally identifiable information is collected.
Since the beginning of the global pandemic of COVID-19, the testing of COVID-19 has remained insufficient. Many COVID-19 symptoms-tracking apps were developed to gain real-time insights about the pandemic to help healthcare professionals and policymakers. However, these apps pit public health against civil liberties, reminding us of the famous question “Do you give up a little liberty to get a little protection?”
COVID Nearby was developed to resolve this tension – with the belief that the growing pandemic is not an excuse to ignore privacy and civil liberties. To do this, experts from epidemiology, communications, privacy, and security collaborated to answer questions such as, “What is the right information to collect?”, “What is privacy in this context?”, and importantly, “How to communicate useful information without hurting the privacy of individuals?”
The innumerable privacy breaches, reported in past years, clearly show that ad-hoc protection mechanisms fail to preserve privacy. So today, the only acceptable privacy guarantee should be a formal privacy guarantee. Thus, COVID Nearby safeguards privacy of all the data contributors via differential privacy, a well-accepted model that provides a mathematical privacy guarantee to all users. Today, differential privacy (DP) is being used by industry (for example, Apple and Google both use DP in their products) and also by government. Most recently, the 2020 US Census used differential privacy to provide privacy. Developing COVID Nearby required solving many unique challenges, e.g., supporting an arbitrary number of queries over the continually updating data with strong correlations. Apart from the technical challenges, the app development process was also quite intensive: they needed to deal with the issues such as the appropriate placement of the information, the intuitiveness of the interface, the need for a consistent cross-platform experience, while educating public about the risk to their privacy as well.