HKBU Helps Contain COVID-19 with AI-enhanced Technology Adopted by LeaveHomeSafe

Aug 13, 2021
Professor Guo Yike (2nd left), Vice-President (Research and Development) of HKBU, Professor Xu Jianliang (1st left), Associate Head and Professor of the Department of Computer Science at HKBU, Mr. Kingsley Wong (2nd right), Assistant Government Chief Information Officer (Industry Development) of the Office of the Government Chief Information Officer (OGCIO) and Mr Brian Sun (1st right), Senior Systems Manager (Industry Development) of OGCIO introduced the dynamic auto-leave function in LeaveHomeSafe mobile app in media briefing.
The display of HKBU’s AI-enhanced dynamic auto-leave function in LeaveHomeSafe mobile app


Digital contact tracing has been playing an instrumental role during the COVID-19 pandemic, being a way to slow down the spread of the highly contagious virus. The privacy-preserving artificial intelligence (AI)-enhanced technology developed by the computer scientists at Hong Kong Baptist University (HKBU) for BU-Trace — the city’s first and the university’s own contact tracing system, is now adopted by the latest version of LeaveHomeSafe, a digital contact tracing mobile app launched by the HKSAR government.

Dynamic auto-leave function enhances user experience

Users of LeaveHomeSafe can now enable the newly added dynamic auto-leave function — an innovative AI-enhanced technology developed by a research team led by Professor Xu Jianliang, Associate Head and Professor of the Department of Computer Science at HKBU — to automatically record the leaving event when getting off a taxi.

Professor Guo Yike, Vice-President (Research & Development) of HKBU, said, “we are honoured that we are making an impact in the community by applying our research and innovations, playing a role in the concerted response to the COVID-19 pandemic which is a significant public health threat. Such new technology provides more accurate contact tracing records and helps fight against the virus.”

Powered by a novel AI algorithm and by utilizing the built-in accelerometer of the mobile phones, the dynamic auto-leave function improves both user experience and the accuracy of the check-out times recorded. During a taxi ride, without the need of tapping on the screen, the system can automatically record user’s leaving time by monitoring the motion patterns detected by the accelerometer.

The function does not need to request for location permission on the mobile devices, and the collected sensor data will not be stored in the system, alleviating users’ privacy concerns.

Award-winning privacy-preserving technologies ease privacy concerns

Despite widely recognized as an effective non-pharmaceutical intervention to identify and isolate infected individuals, digital contact tracing has raised privacy concerns. Professor Xu and his team successfully came up with groundbreaking privacy-preserving solutions which were first integrated with the university’s COVID-19 alert system, BU-Trace.

Professor Xu said, “With the use of novel AI technologies, we developed new solutions to upgrade our COVID-19 alert system 2.0, making it more user-friendly and eliminating users’ privacy concerns. That encouraged more members of the University to use the alert system, boosting the effectiveness of the system.”

BU-Trace, implemented on the HKBU campus since October 2020 to enhance its COVID-19 risks alert capability, has several privacy-preserving features to protect users’ privacy. The system was awarded “The Best Paper Award” by International Workshop on Mobile Ubiquitous Systems and Technologies earlier this year and was recently shortlisted for Technological or Digital Innovation of the Year in the Times Higher Education (THE) Awards Asia 2021.

No personal and location data collected by the system

Leveraging permissionless QR-code and NFC (near field communication) technology to record users’ venue check-in information and taking a decentralized approach, BU-Trace does not ask for location permission or collect any personal information and location data, while all venue check-in records are only stored on the devices. Such features enable users to confidentially conduct contact matching, safeguarding user’s privacy and making the system more scalable.

Launched in November 2020 by the government, LeaveHomeSafe mobile app is a digital tool for recording user’s visit history of different venues and taxi rides, and to provide COVID-19 exposure notification. As of today, the mobile app has been downloaded over 5 million times, with about 92,000 venues and 18,000 taxis in the city participating and displaying LeaveHomeSafe QR codes.